Split (1)

train · 10k rows

url stringlengths 63 63	repository_url stringclasses 1 value	labels_url stringlengths 77 77	comments_url stringlengths 72 72	events_url stringlengths 70 70	html_url stringlengths 51 53	id int64 1.57B 2.35B	node_id stringlengths 18 19	number int64 59.5k 69.6k	title stringlengths 1 554	user dict	labels listlengths 0 8	state stringclasses 2 values	locked bool 2 classes	assignee dict	assignees listlengths 0 8	milestone null	comments listlengths 0 30	created_at timestamp[s]	updated_at timestamp[s]	closed_at timestamp[s]	author_association stringclasses 4 values	active_lock_reason stringclasses 3 values	draft bool 2 classes	pull_request dict	body stringlengths 1 65.4k ⌀	reactions dict	timeline_url stringlengths 72 72	performed_via_github_app null	state_reason stringclasses 3 values	is_pull_request bool 2 classes
https://api.github.com/repos/tensorflow/tensorflow/issues/62483	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62483/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62483/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62483/events	https://github.com/tensorflow/tensorflow/issues/62483	2,012,089,617	I_kwDOArmXAs537g0R	62,483	Multi-GPU strategy scope causes save_weights error	{ "login": "moberweger", "id": 10559836, "node_id": "MDQ6VXNlcjEwNTU5ODM2", "avatar_url": "https://avatars.githubusercontent.com/u/10559836?v=4", "gravatar_id": "", "url": "https://api.github.com/users/moberweger", "html_url": "https://github.com/moberweger", "followers_url": "https://api.github.com/users/moberweger/followers", "following_url": "https://api.github.com/users/moberweger/following{/other_user}", "gists_url": "https://api.github.com/users/moberweger/gists{/gist_id}", "starred_url": "https://api.github.com/users/moberweger/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/moberweger/subscriptions", "organizations_url": "https://api.github.com/users/moberweger/orgs", "repos_url": "https://api.github.com/users/moberweger/repos", "events_url": "https://api.github.com/users/moberweger/events{/privacy}", "received_events_url": "https://api.github.com/users/moberweger/received_events", "type": "User", "site_admin": false }	[ { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547538, "node_id": "MDU6TGFiZWwxMDk3NTQ3NTM4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:gpu", "name": "comp:gpu", "color": "0052cc", "default": false, "description": "GPU related issues" }, { "id": 5508003926, "node_id": "LA_kwDOArmXAs8AAAABSE14Vg", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.13", "name": "TF 2.13", "color": "B13ACB", "default": false, "description": "For issues related to Tensorflow 2.13" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @moberweger ,\r\n\r\nI have replicated the reported behaviour with colab using TF v2.14, 2.15. Please find the [gist](https://colab.research.google.com/gist/Venkat6871/24ca1370e913df5ba221d609d8b0385a/62483_gpu-2-14-2-15.ipynb) here for reference.\r\n\r\nThank you!", "Hi @moberweger ,\r\n\r\nI have done the code changes to the attached code to make it compatible to save and load in `.keras` format. \r\n\r\nSince there are custom objects in the model we need to explicitly override `get_config()` and `from_config()` methods. Please find the gists that executes fine with [keras3 ](https://colab.research.google.com/gist/SuryanarayanaY/a21c8f3b4327f4a41abc396af303df5b/62483_gpu_r1-keras3.ipynb)and [kears-2.15](https://colab.research.google.com/gist/SuryanarayanaY/577552b767bcdd5ecaaec35bedcc89eb/62483_gpu_r1-keras-2-15v.ipynb) as well.\r\n\r\nThanks!", "Thanks @SuryanarayanaY for looking into this.\r\nI checked and the specific code works on my multi-GPU setup, but I observed some inconsistencies: \r\n1. It seems that the extension `.hdf5` vs `.h5` has an impact, since it still fails for `.hdf5` file format but it works on `.h5`\r\n2. The gist for `keras-2.15` outputs `0 conv2d/kernel:0` whereas `keras-3.0` outputs `0 kernel`\r\n3. It is interesting to see that it also works without the `.keras` format\r\n\r\nThanks!" ]	2023-11-27T11:41:57	2024-01-22T13:50:29	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source binary ### TensorFlow version 2.13.1 ### Custom code Yes ### OS platform and distribution Ubuntu 20.04 ### Mobile device _No response_ ### Python version 3.8 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version 11.8 ### GPU model and memory 2x A6000 48G ### Current behavior? I have a custom `Layer` with a `Model` that is created and saved within a multi-GPU scope. Everything works well without the `MirroredStrategy` scope, thus running a single GPU. But when running on multiple GPUs using the strategy scope, the names of the weights of the model get truncated and essentially only resemble `kernel` and `bias`. This leads to duplicate entries in the `save_weights` function causing an HDF5 error. Code to reproduce is below (you need to have >=2 GPUs), and log output is also attached. ### Standalone code to reproduce the issue ```shell import os import shutil import tensorflow as tf from contextlib import nullcontext from tensorflow.keras.layers import Input, Conv2D, Layer, GroupNormalization from tensorflow.keras.models import Model, load_model class MyLayer(Layer): def __init__(self, channels, kwargs): super().__init__(kwargs) self.norm = GroupNormalization(epsilon=1e-5) self.proj1 = Conv2D(channels, 1) self.proj2 = Conv2D(channels, 1) def call(self, inputs): return self.proj2(self.proj1(self.norm(inputs))) class ModelMock(Model): def __init__(self, img_height, img_width, name=None): x_input = Input((img_height, img_width, 3), name="x_input") x = Conv2D(320, kernel_size=3, padding="same")(x_input) x = MyLayer(320)(x) output = Conv2D(16, kernel_size=3, padding="same")(x) super().__init__([x_input], output, name=name) if __name__ == '__main__': strategy = tf.distribute.MirroredStrategy() print("Number of devices: {}".format(strategy.num_replicas_in_sync)) if os.path.exists("./test"): shutil.rmtree("./test/") os.mkdir("./test/") fail = True with strategy.scope() if fail else nullcontext(): print("CREATING") model = ModelMock(256, 256) for i, w in enumerate(model.weights): print(i, w.name) model.save_weights("./test/model1.hdf5") model.save("./test/model1") model = load_model("./test/model1", compile=False) for i, w in enumerate(model.weights): print(i, w.name) model.save_weights("./test/model2.hdf5") print("DONE") ``` ### Relevant log output ```shell Number of devices: 2 CREATING 0 conv2d/kernel:0 1 conv2d/bias:0 2 my_layer/group_normalization/gamma:0 3 my_layer/group_normalization/beta:0 4 my_layer/conv2d_1/kernel:0 5 my_layer/conv2d_1/bias:0 6 my_layer/conv2d_2/kernel:0 7 my_layer/conv2d_2/bias:0 8 conv2d_3/kernel:0 9 conv2d_3/bias:0 0 kernel:0 1 bias:0 2 gamma:0 3 beta:0 4 kernel:0 5 bias:0 6 kernel:0 7 bias:0 8 kernel:0 9 bias:0 Traceback (most recent call last): File "/home/naked/dev/meta-cv/nak3d-fusion/python/apps/ml/check_weight_names.py", line 48, in <module> model.save_weights("./test/model2.hdf5") File "/usr/local/lib/python3.8/dist-packages/keras/src/utils/traceback_utils.py", line 70, in error_handler raise e.with_traceback(filtered_tb) from None File "/usr/local/lib/python3.8/dist-packages/h5py/_hl/group.py", line 183, in create_dataset dsid = dataset.make_new_dset(group, shape, dtype, data, name, **kwds) File "/usr/local/lib/python3.8/dist-packages/h5py/_hl/dataset.py", line 163, in make_new_dset dset_id = h5d.create(parent.id, name, tid, sid, dcpl=dcpl, dapl=dapl) File "h5py/_objects.pyx", line 54, in h5py._objects.with_phil.wrapper File "h5py/_objects.pyx", line 55, in h5py._objects.with_phil.wrapper File "h5py/h5d.pyx", line 138, in h5py.h5d.create ValueError: Unable to create dataset (name already exists) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62483/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62483/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62482	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62482/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62482/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62482/events	https://github.com/tensorflow/tensorflow/pull/62482	2,011,829,573	PR_kwDOArmXAs5gaj7b	62,482	[oneDNN] Dynamic generate oneDNN version	{ "login": "Zantares", "id": 38638514, "node_id": "MDQ6VXNlcjM4NjM4NTE0", "avatar_url": "https://avatars.githubusercontent.com/u/38638514?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Zantares", "html_url": "https://github.com/Zantares", "followers_url": "https://api.github.com/users/Zantares/followers", "following_url": "https://api.github.com/users/Zantares/following{/other_user}", "gists_url": "https://api.github.com/users/Zantares/gists{/gist_id}", "starred_url": "https://api.github.com/users/Zantares/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Zantares/subscriptions", "organizations_url": "https://api.github.com/users/Zantares/orgs", "repos_url": "https://api.github.com/users/Zantares/repos", "events_url": "https://api.github.com/users/Zantares/events{/privacy}", "received_events_url": "https://api.github.com/users/Zantares/received_events", "type": "User", "site_admin": false }	[ { "id": 1104829434, "node_id": "MDU6TGFiZWwxMTA0ODI5NDM0", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:mkl", "name": "comp:mkl", "color": "0052cc", "default": false, "description": "MKL related issues" }, { "id": 1169365494, "node_id": "MDU6TGFiZWwxMTY5MzY1NDk0", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:M", "name": "size:M", "color": "adafea", "default": false, "description": "CL Change Size: Medium" } ]	closed	false	{ "login": "penpornk", "id": 38085909, "node_id": "MDQ6VXNlcjM4MDg1OTA5", "avatar_url": "https://avatars.githubusercontent.com/u/38085909?v=4", "gravatar_id": "", "url": "https://api.github.com/users/penpornk", "html_url": "https://github.com/penpornk", "followers_url": "https://api.github.com/users/penpornk/followers", "following_url": "https://api.github.com/users/penpornk/following{/other_user}", "gists_url": "https://api.github.com/users/penpornk/gists{/gist_id}", "starred_url": "https://api.github.com/users/penpornk/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/penpornk/subscriptions", "organizations_url": "https://api.github.com/users/penpornk/orgs", "repos_url": "https://api.github.com/users/penpornk/repos", "events_url": "https://api.github.com/users/penpornk/events{/privacy}", "received_events_url": "https://api.github.com/users/penpornk/received_events", "type": "User", "site_admin": false }	[ { "login": "penpornk", "id": 38085909, "node_id": "MDQ6VXNlcjM4MDg1OTA5", "avatar_url": "https://avatars.githubusercontent.com/u/38085909?v=4", "gravatar_id": "", "url": "https://api.github.com/users/penpornk", "html_url": "https://github.com/penpornk", "followers_url": "https://api.github.com/users/penpornk/followers", "following_url": "https://api.github.com/users/penpornk/following{/other_user}", "gists_url": "https://api.github.com/users/penpornk/gists{/gist_id}", "starred_url": "https://api.github.com/users/penpornk/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/penpornk/subscriptions", "organizations_url": "https://api.github.com/users/penpornk/orgs", "repos_url": "https://api.github.com/users/penpornk/repos", "events_url": "https://api.github.com/users/penpornk/events{/privacy}", "received_events_url": "https://api.github.com/users/penpornk/received_events", "type": "User", "site_admin": false }, { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "This PR is built based on an old commit and got failure in the newest master, I will check and address it.", "Hi @penpornk Can you please review this PR ? Thank you!\r\n", "Hi @penpornk Can you please review this PR ? Thank you!", "Hi @Zantares Can you please rebase your branch and resolve conflicts? Thank you!", "This PR is stale because it has been open for 14 days with no activity. It will be closed if no further activity occurs. Thank you.", "Hi @Zantares Can you please rebase your branch and resolve conflicts? Thank you!", "Hi @Zantares Can you please rebase your branch and resolve conflicts? Thank you!", "It seems this method only works on bare metal and failed in docker. I will have a check. ", "Hi @Zantares Can you please rebase your branch and resolve conflicts? Thank you!", "I'm sorry for that I didn't get time to debug the docker issue. I plan to close this PR first and resubmit it in the future once the docker issue is solved, thanks!" ]	2023-11-27T09:12:23	2024-05-29T05:30:03	2024-05-29T05:30:03	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62482", "html_url": "https://github.com/tensorflow/tensorflow/pull/62482", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62482.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62482.patch", "merged_at": null }	Added a python script to get oneDNN version from cmake and dynamic set it to bazel, instead of changing the oneDNN version manually.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62482/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62482/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62481	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62481/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62481/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62481/events	https://github.com/tensorflow/tensorflow/issues/62481	2,011,767,912	I_kwDOArmXAs536SRo	62,481	Discrepancy in TensorFlow XLA Compiled Models Due to Different Multiplication Orders with `tf.abs` on GPU	{ "login": "Gwihwan-Go", "id": 83144588, "node_id": "MDQ6VXNlcjgzMTQ0NTg4", "avatar_url": "https://avatars.githubusercontent.com/u/83144588?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Gwihwan-Go", "html_url": "https://github.com/Gwihwan-Go", "followers_url": "https://api.github.com/users/Gwihwan-Go/followers", "following_url": "https://api.github.com/users/Gwihwan-Go/following{/other_user}", "gists_url": "https://api.github.com/users/Gwihwan-Go/gists{/gist_id}", "starred_url": "https://api.github.com/users/Gwihwan-Go/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Gwihwan-Go/subscriptions", "organizations_url": "https://api.github.com/users/Gwihwan-Go/orgs", "repos_url": "https://api.github.com/users/Gwihwan-Go/repos", "events_url": "https://api.github.com/users/Gwihwan-Go/events{/privacy}", "received_events_url": "https://api.github.com/users/Gwihwan-Go/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1133285679, "node_id": "MDU6TGFiZWwxMTMzMjg1Njc5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:xla", "name": "comp:xla", "color": "0052cc", "default": false, "description": "XLA" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	open	false	{ "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false }	[ { "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false } ]	null	[ "Here is another piece of code that triggers the same error. I hope this assists you in identifying the root cause of the bug\r\n```python\r\nfrom typing import Dict\r\nimport tensorflow as tf\r\nimport pickle\r\nimport os\r\nimport numpy as np\r\n\r\nclass Model1(tf.keras.Model):\r\n def __init__(self):\r\n super().__init__()\r\n\r\n @tf.function(jit_compile=True)\r\n def __call__(self, inp1, inp2):\r\n # Forward pass logic using TensorFlow operations\r\n _abs = tf.abs(tf.multiply(tf.multiply(inp1, inp2), tf.multiply(inp1, inp2)))\r\n return _abs\r\n\r\nclass Model2(tf.keras.Model):\r\n def __init__(self):\r\n super().__init__()\r\n\r\n @tf.function(jit_compile=True)\r\n def __call__(self, inp1, inp2):\r\n # Forward pass logic using TensorFlow operations\r\n _abs = tf.abs(tf.multiply(inp1, tf.multiply(inp2, tf.multiply(inp1, inp2))))\r\n return _abs\r\n\r\ninputs = [\r\ntf.cast(tf.random.uniform(shape=[11], minval=-128, maxval=128, dtype=tf.int32), tf.int16),\r\ntf.cast(tf.random.uniform(shape=[], minval=-128, maxval=128, dtype=tf.int32), tf.int16),\r\n]\r\nmodel1 = Model1()\r\nmodel2 = Model2()\r\ndevice = \"gpu\"\r\nwith tf.device(device):\r\n tf.config.run_functions_eagerly(True)\r\n out1 = model1(inputs)\r\n out2 = model2(inputs)\r\n print(f'=========eager_output(version:{tf.__version__})================')\r\n try :\r\n for i in range(min(len(out1),len(out2))):\r\n np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th')\r\n print(\"XLA_eager does not trigger assertion\")\r\n except AssertionError as e:\r\n print(\"XLA_eager triggers assertion\")\r\n print(e)\r\n tf.config.run_functions_eagerly(False)\r\n out1 = model1(inputs)\r\n out2 = model2(inputs)\r\n print(f'=========compiled_output(version:{tf.__version__})================')\r\n try :\r\n for i in range(min(len(out1),len(out2))):\r\n np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th')\r\n print(\"XLA_complie does not trigger assertion\")\r\n except AssertionError as e:\r\n print(\"XLA_complie triggers assertion\")\r\n print(e)\r\n```\r\nand log output : \r\n```\r\n=========compiled_output(version:2.15.0)================\r\nXLA_complie triggers assertion\r\n\r\nNot equal to tolerance rtol=0.001, atol=0.001\r\nat checking 0th\r\nMismatched elements: 1 / 1 (100%)\r\nMax absolute difference: 1736\r\nMax relative difference: 0.05442006\r\n x: array(-31900, dtype=int16)\r\n y: array(31900, dtype=int16)\r\n```", "@tilakrayal Hey, have you reproduced this issue?" ]	2023-11-27T08:36:40	2023-12-05T04:08:35	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.15.0 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04.3 LTS ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version cuda : 12.2 / cudnn 8.9.04 ### GPU model and memory Tesla V100S-PCIE-32GB ### Current behavior? In TensorFlow 2.15.0, I've encountered an issue where two mathematically equivalent models yield different results when compiled with XLA. The discrepancy appears to be tied to the order of multiplication operations. This issue is critical as it questions the consistency and reliability of TensorFlow's computation, especially in a compiled environment. The models yield dramatically different results under XLA compilation, with discrepancies that are far from trivial. ### Standalone code to reproduce the issue ```python import tensorflow as tf import os import numpy as np class Model1(tf.keras.Model): @tf.function(jit_compile=True) def __call__(self, inp1, inp2, inp3): # inp3(inp2inp1)(inp3(inp2inp1)) mul3 = tf.multiply(tf.multiply(inp3, tf.multiply(inp1, inp2)), tf.multiply(inp3, tf.multiply(inp1, inp2))) _abs = tf.abs(mul3) return mul3, _abs class Model2(tf.keras.Model): @tf.function(jit_compile=True) def __call__(self, inp1, inp2, inp3): # inp3 ((inp1inp2)(inp3(inp1inp2))) mul3 = tf.multiply(inp3, tf.multiply(tf.multiply(inp1, inp2), tf.multiply(inp3, tf.multiply(inp1, inp2)))) _abs = tf.abs(mul3) return mul3, _abs with tf.device(tf.config.list_logical_devices('GPU')[0].name): inputs = [ tf.random.uniform(shape=[3], minval=-100, maxval=100, dtype=tf.int32), tf.random.uniform(shape=[], minval=-100, maxval=100, dtype=tf.int32), tf.random.uniform(shape=[3], minval=-100, maxval=100, dtype=tf.int32), ] model1 = Model1() model2 = Model2() tf.config.run_functions_eagerly(True) out1 = model1(inputs) out2 = model2(inputs) print(f'=========eager_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th') print("XLA_eager does not trigger assertion") except AssertionError as e: print("XLA_eager triggers assertion") print(e) tf.config.run_functions_eagerly(False) out1 = model1(inputs) out2 = model2(inputs) print(f'=========compiled_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th') print("XLA_complie does not trigger assertion") except AssertionError as e: print("XLA_complie triggers assertion") print(e) ``` ### Relevant log output ```shell 2023-11-27 08:25:14.134072: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered 2023-11-27 08:25:14.134128: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered 2023-11-27 08:25:14.135901: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered 2023-11-27 08:25:14.142938: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. 2023-11-27 08:25:14.955654: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT 2023-11-27 08:25:25.751708: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 1123 MB memory: -> device: 0, name: Tesla V100S-PCIE-32GB, pci bus id: 0000:01:00.0, compute capability: 7.0 2023-11-27 08:25:25.754126: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:1 with 15774 MB memory: -> device: 1, name: Tesla V100S-PCIE-32GB, pci bus id: 0000:24:00.0, compute capability: 7.0 2023-11-27 08:25:25.755764: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:2 with 16176 MB memory: -> device: 2, name: Tesla V100S-PCIE-32GB, pci bus id: 0000:41:00.0, compute capability: 7.0 2023-11-27 08:25:25.757607: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:3 with 17872 MB memory: -> device: 3, name: Tesla V100S-PCIE-32GB, pci bus id: 0000:61:00.0, compute capability: 7.0 2023-11-27 08:25:25.759556: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:4 with 18494 MB memory: -> device: 4, name: Tesla V100S-PCIE-32GB, pci bus id: 0000:81:00.0, compute capability: 7.0 2023-11-27 08:25:25.761281: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:5 with 19166 MB memory: -> device: 5, name: Tesla V100S-PCIE-32GB, pci bus id: 0000:a1:00.0, compute capability: 7.0 2023-11-27 08:25:25.763252: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:6 with 20272 MB memory: -> device: 6, name: Tesla V100S-PCIE-32GB, pci bus id: 0000:c1:00.0, compute capability: 7.0 2023-11-27 08:25:25.765351: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:7 with 20720 MB memory: -> device: 7, name: Tesla V100S-PCIE-32GB, pci bus id: 0000:e1:00.0, compute capability: 7.0 =========eager_output(version:2.15.0)================ XLA_eager does not trigger assertion 2023-11-27 08:25:26.220882: I external/local_xla/xla/service/service.cc:168] XLA service 0x555a544232d0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: 2023-11-27 08:25:26.220929: I external/local_xla/xla/service/service.cc:176] StreamExecutor device (0): Tesla V100S-PCIE-32GB, Compute Capability 7.0 rvice.cc:176] 2023-11-27 08:25:26.226472: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:269] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable. 2023-11-27 08:25:26.908888: I external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:454] Loaded cuDNN version 8904 WARNING: All log messages before absl::InitializeLog() is called are written to STDERR I0000 00:00:1701073527.016479 336904 device_compiler.h:186] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process. =========compiled_output(version:2.15.0)================ XLA_complie triggers assertion Not equal to tolerance rtol=0.001, atol=0.001 at checking 1th # indicates error happens after tf.abs Mismatched elements: 1 / 3 (33.3%) Max absolute difference: 1964494976 Max relative difference: 1.68592003 x: array([-1165236160, 395214400, 1192464], dtype=int32) y: array([1165236160, 395214400, 1192464], dtype=int32) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62481/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62481/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62480	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62480/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62480/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62480/events	https://github.com/tensorflow/tensorflow/issues/62480	2,011,297,081	I_kwDOArmXAs534fU5	62,480	No gradients provided for any variable	{ "login": "dalalkrish", "id": 15665826, "node_id": "MDQ6VXNlcjE1NjY1ODI2", "avatar_url": "https://avatars.githubusercontent.com/u/15665826?v=4", "gravatar_id": "", "url": "https://api.github.com/users/dalalkrish", "html_url": "https://github.com/dalalkrish", "followers_url": "https://api.github.com/users/dalalkrish/followers", "following_url": "https://api.github.com/users/dalalkrish/following{/other_user}", "gists_url": "https://api.github.com/users/dalalkrish/gists{/gist_id}", "starred_url": "https://api.github.com/users/dalalkrish/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/dalalkrish/subscriptions", "organizations_url": "https://api.github.com/users/dalalkrish/orgs", "repos_url": "https://api.github.com/users/dalalkrish/repos", "events_url": "https://api.github.com/users/dalalkrish/events{/privacy}", "received_events_url": "https://api.github.com/users/dalalkrish/received_events", "type": "User", "site_admin": false }	[ { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62480\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62480\">No</a>\n", "After carefully reading through other similar issues, I found out that if the final output layer or anywhere in the graph had any TF functions it won't calculate gradient. I had `tf.abs` and `tf.round` in my final layer which was preventing gradient calculation and throwing this error. After, removing them the model trains as expected" ]	2023-11-27T00:50:06	2023-11-27T01:02:13	2023-11-27T00:59:03	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.15.0 ### Custom code Yes ### OS platform and distribution _No response_ ### Mobile device _No response_ ### Python version 3.10.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? When calling `fit` on sub-classed Layers and Model, it throws ValueError: No gradients provided for any variable with or without Gradient Tape context. I have also tried defining functions with sub-layers and constructing model using Model API i.e. not sub-classing it but I get the same error ### Standalone code to reproduce the issue ```shell I'm putting a link to the question I have asked on Stackoverflow as it is quite a bit of code to reproduce the issue: https://stackoverflow.com/questions/77546477/how-to-build-a-model-and-train-it-with-tensorflow-keras-sub-classing ``` ### Relevant log output ```shell File "/local_disk0/.ephemeral_nfs/cluster_libraries/python/lib/python3.10/site-packages/keras/src/engine/training.py", line 1401, in train_function * return step_function(self, iterator) File "/local_disk0/.ephemeral_nfs/cluster_libraries/python/lib/python3.10/site-packages/keras/src/engine/training.py", line 1384, in step_function outputs = model.distribute_strategy.run(run_step, args=(data,)) File "/local_disk0/.ephemeral_nfs/cluster_libraries/python/lib/python3.10/site-packages/keras/src/engine/training.py", line 1373, in run_step outputs = model.train_step(data) File "/root/.ipykernel/2759/command-461111845465809-3511162167", line 27, in train_step self.optimizer.apply_gradients(zip(gradients, trainable_vars)) File "/local_disk0/.ephemeral_nfs/cluster_libraries/python/lib/python3.10/site-packages/keras/src/optimizers/optimizer.py", line 1222, in apply_gradients grads_and_vars = self.aggregate_gradients(grads_and_vars) File "/local_disk0/.ephemeral_nfs/cluster_libraries/python/lib/python3.10/site-packages/keras/src/optimizers/optimizer.py", line 1184, in aggregate_gradients return optimizer_utils.all_reduce_sum_gradients(grads_and_vars) File "/local_disk0/.ephemeral_nfs/cluster_libraries/python/lib/python3.10/site-packages/keras/src/optimizers/utils.py", line 33, in all_reduce_sum_gradients filtered_grads_and_vars = filter_empty_gradients(grads_and_vars) File "/local_disk0/.ephemeral_nfs/cluster_libraries/python/lib/python3.10/site-packages/keras/src/optimizers/utils.py", line 77, in filter_empty_gradients raise ValueError( ValueError: No gradients provided for any variable: (['trxster_7/encoder_10/en_embed_layer/embeddings:0', 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/query/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/query/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/key/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/key/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/value/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/value/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/attention_output/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/attention_output/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_136/dense_283/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_136/dense_283/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_136/dense_284/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_136/dense_284/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_136/layer_normalization_340/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_136/layer_normalization_340/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_136/layer_normalization_341/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_136/layer_normalization_341/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/query/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/query/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/key/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/key/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/value/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/value/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/attention_output/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/attention_output/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_137/dense_285/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_137/dense_285/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_137/dense_286/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_137/dense_286/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_137/layer_normalization_342/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_137/layer_normalization_342/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_137/layer_normalization_343/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_137/layer_normalization_343/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/query/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/query/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/key/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/key/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/value/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/value/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/attention_output/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/attention_output/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_138/dense_287/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_138/dense_287/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_138/dense_288/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_138/dense_288/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_138/layer_normalization_344/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_138/layer_normalization_344/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_138/layer_normalization_345/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_138/layer_normalization_345/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/query/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/query/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/key/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/key/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/value/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/value/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/attention_output/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/attention_output/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_139/dense_289/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_139/dense_289/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_139/dense_290/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_139/dense_290/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_139/layer_normalization_346/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_139/layer_normalization_346/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_139/layer_normalization_347/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_139/layer_normalization_347/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/query/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/query/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/key/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/key/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/value/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/value/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/attention_output/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/attention_output/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_140/dense_291/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_140/dense_291/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_140/dense_292/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_140/dense_292/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_140/layer_normalization_348/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_140/layer_normalization_348/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_140/layer_normalization_349/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_140/layer_normalization_349/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/query/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/query/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/key/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/key/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/value/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/value/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/attention_output/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/attention_output/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_141/dense_293/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_141/dense_293/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_141/dense_294/kernel:0', 'trxster_7/encoder_10/encoder_sub_layer_141/dense_294/bias:0', 'trxster_7/encoder_10/encoder_sub_layer_141/layer_normalization_350/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_141/layer_normalization_350/beta:0', 'trxster_7/encoder_10/encoder_sub_layer_141/layer_normalization_351/gamma:0', 'trxster_7/encoder_10/encoder_sub_layer_141/layer_normalization_351/beta:0', 'trxster_7/decoder_11/de_embed_layer/embeddings:0', 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/dense_295/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/dense_295/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/dense_296/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/dense_296/bias:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/layer_normalization_352/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/layer_normalization_352/beta:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/layer_normalization_353/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/layer_normalization_353/beta:0', 'trxster_7/decoder_11/decoder_sublayer_69/layer_normalization_354/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_69/layer_normalization_354/beta:0', 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/dense_297/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/dense_297/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/dense_298/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/dense_298/bias:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/layer_normalization_355/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/layer_normalization_355/beta:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/layer_normalization_356/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/layer_normalization_356/beta:0', 'trxster_7/decoder_11/decoder_sublayer_70/layer_normalization_357/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_70/layer_normalization_357/beta:0', 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/dense_299/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/dense_299/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/dense_300/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/dense_300/bias:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/layer_normalization_358/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/layer_normalization_358/beta:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/layer_normalization_359/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/layer_normalization_359/beta:0', 'trxster_7/decoder_11/decoder_sublayer_71/layer_normalization_360/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_71/layer_normalization_360/beta:0', 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/dense_301/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/dense_301/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/dense_302/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/dense_302/bias:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/layer_normalization_361/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/layer_normalization_361/beta:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/layer_normalization_362/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/layer_normalization_362/beta:0', 'trxster_7/decoder_11/decoder_sublayer_72/layer_normalization_363/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_72/layer_normalization_363/beta:0', 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/dense_303/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/dense_303/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/dense_304/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/dense_304/bias:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/layer_normalization_364/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/layer_normalization_364/beta:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/layer_normalization_365/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/layer_normalization_365/beta:0', 'trxster_7/decoder_11/decoder_sublayer_73/layer_normalization_366/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_73/layer_normalization_366/beta:0', 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/query/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/query/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/key/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/key/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/value/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/value/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/attention_output/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/attention_output/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/dense_305/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/dense_305/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/dense_306/kernel:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/dense_306/bias:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/layer_normalization_367/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/layer_normalization_367/beta:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/layer_normalization_368/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/layer_normalization_368/beta:0', 'trxster_7/decoder_11/decoder_sublayer_74/layer_normalization_369/gamma:0', 'trxster_7/decoder_11/decoder_sublayer_74/layer_normalization_369/beta:0', 'trxster_7/decoder_11/output_layer/kernel:0', 'trxster_7/decoder_11/output_layer/bias:0'],). Provided `grads_and_vars` is ((None, <tf.Variable 'trxster_7/encoder_10/en_embed_layer/embeddings:0' shape=(500, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/en_mha_layer_1/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/dense_283/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/dense_283/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/dense_284/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/dense_284/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/layer_normalization_340/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/layer_normalization_340/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/layer_normalization_341/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_136/layer_normalization_341/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/en_mha_layer_2/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/dense_285/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/dense_285/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/dense_286/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/dense_286/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/layer_normalization_342/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/layer_normalization_342/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/layer_normalization_343/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_137/layer_normalization_343/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/en_mha_layer_3/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/dense_287/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/dense_287/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/dense_288/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/dense_288/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/layer_normalization_344/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/layer_normalization_344/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/layer_normalization_345/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_138/layer_normalization_345/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/en_mha_layer_4/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/dense_289/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/dense_289/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/dense_290/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/dense_290/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/layer_normalization_346/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/layer_normalization_346/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/layer_normalization_347/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_139/layer_normalization_347/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/en_mha_layer_5/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/dense_291/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/dense_291/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/dense_292/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/dense_292/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/layer_normalization_348/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/layer_normalization_348/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/layer_normalization_349/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_140/layer_normalization_349/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/en_mha_layer_6/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/dense_293/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/dense_293/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/dense_294/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/dense_294/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/layer_normalization_350/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/layer_normalization_350/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/layer_normalization_351/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/encoder_10/encoder_sub_layer_141/layer_normalization_351/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/de_embed_layer/embeddings:0' shape=(500, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/de_mha_layer_1/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/de_mha_layer_1/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/dense_295/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/dense_295/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/dense_296/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/dense_296/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/layer_normalization_352/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/layer_normalization_352/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/layer_normalization_353/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/encoder_sub_layer_142/layer_normalization_353/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/layer_normalization_354/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_69/layer_normalization_354/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/de_mha_layer_2/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/de_mha_layer_2/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/dense_297/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/dense_297/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/dense_298/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/dense_298/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/layer_normalization_355/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/layer_normalization_355/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/layer_normalization_356/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/encoder_sub_layer_143/layer_normalization_356/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/layer_normalization_357/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_70/layer_normalization_357/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/de_mha_layer_3/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/de_mha_layer_3/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/dense_299/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/dense_299/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/dense_300/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/dense_300/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/layer_normalization_358/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/layer_normalization_358/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/layer_normalization_359/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/encoder_sub_layer_144/layer_normalization_359/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/layer_normalization_360/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_71/layer_normalization_360/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/de_mha_layer_4/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/de_mha_layer_4/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/dense_301/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/dense_301/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/dense_302/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/dense_302/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/layer_normalization_361/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/layer_normalization_361/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/layer_normalization_362/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/encoder_sub_layer_145/layer_normalization_362/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/layer_normalization_363/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_72/layer_normalization_363/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/de_mha_layer_5/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/de_mha_layer_5/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/dense_303/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/dense_303/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/dense_304/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/dense_304/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/layer_normalization_364/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/layer_normalization_364/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/layer_normalization_365/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/encoder_sub_layer_146/layer_normalization_365/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/layer_normalization_366/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_73/layer_normalization_366/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/de_mha_layer_6/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/query/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/query/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/key/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/key/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/value/kernel:0' shape=(512, 8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/value/bias:0' shape=(8, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/attention_output/kernel:0' shape=(8, 512, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/de_mha_layer_6/attention_output/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/dense_305/kernel:0' shape=(512, 2048) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/dense_305/bias:0' shape=(2048,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/dense_306/kernel:0' shape=(2048, 512) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/dense_306/bias:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/layer_normalization_367/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/layer_normalization_367/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/layer_normalization_368/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/encoder_sub_layer_147/layer_normalization_368/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/layer_normalization_369/gamma:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/decoder_sublayer_74/layer_normalization_369/beta:0' shape=(512,) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/output_layer/kernel:0' shape=(512, 1) dtype=float32>), (None, <tf.Variable 'trxster_7/decoder_11/output_layer/bias:0' shape=(1,) dtype=float32>)). ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62480/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62480/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62479	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62479/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62479/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62479/events	https://github.com/tensorflow/tensorflow/issues/62479	2,011,253,512	I_kwDOArmXAs534UsI	62,479	When I import tensorflow as tf with python=3.9.6,there are errors	{ "login": "ABtwo", "id": 33176666, "node_id": "MDQ6VXNlcjMzMTc2NjY2", "avatar_url": "https://avatars.githubusercontent.com/u/33176666?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ABtwo", "html_url": "https://github.com/ABtwo", "followers_url": "https://api.github.com/users/ABtwo/followers", "following_url": "https://api.github.com/users/ABtwo/following{/other_user}", "gists_url": "https://api.github.com/users/ABtwo/gists{/gist_id}", "starred_url": "https://api.github.com/users/ABtwo/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ABtwo/subscriptions", "organizations_url": "https://api.github.com/users/ABtwo/orgs", "repos_url": "https://api.github.com/users/ABtwo/repos", "events_url": "https://api.github.com/users/ABtwo/events{/privacy}", "received_events_url": "https://api.github.com/users/ABtwo/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 3255468475, "node_id": "MDU6TGFiZWwzMjU1NDY4NDc1", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/2.6.0", "name": "2.6.0", "color": "FA96B6", "default": false, "description": "" } ]	closed	false	{ "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false }	[ { "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false } ]	null	[ "hello have u installed tensorflow library", "@ABtwo Please try to upgrade to the latest TF version as you are using an older version. In a virtual environment, you may need to uninstall Python 3.9.6 and install a different version of Python, such as Python 3.8 or Python 3.10. \r\nPlease have a look at the following code to execute the same in virtual environment;\r\n```\r\npython3.8 -m venv tfenv\r\nsource tfenv/bin/activate\r\npip install tensorflow\r\n```\r\n\r\n\r\nThank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "I also met this problem. and i browsered https://numpy.org/devdocs/release/1.20.0-notes.html found that it's obviously tensorflow are using numpy before 2.0\r\nI just change the numpy version pip install numpy==1.16.0 and at least i can use it now ,although it may have some problem", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62479\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62479\">No</a>\n" ]	2023-11-26T23:06:09	2023-12-21T01:48:40	2023-12-21T01:48:37	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version tensorflow-GPU 2.6 ### Custom code Yes ### OS platform and distribution _No response_ ### Mobile device _No response_ ### Python version 3.9 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version 12.3 cudnn version :8.9.6 ### GPU model and memory _No response_ ### Current behavior? import tensorflow as tf C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\framework\dtypes.py:585: FutureWarning: In the future `np.object` will be defined as the corresponding NumPy scalar. np.object, Traceback (most recent call last): File "<stdin>", line 1, in <module> File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\__init__.py", line 41, in <module> from tensorflow.python.tools import module_util as _module_util File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\__init__.py", line 46, in <module> from tensorflow.python import data File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\__init__.py", line 25, in <module> from tensorflow.python.data import experimental File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\experimental\__init__.py", line 97, in <module> from tensorflow.python.data.experimental import service File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\experimental\service\__init__.py", line 353, in <module> from tensorflow.python.data.experimental.ops.data_service_ops import distribute File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\experimental\ops\data_service_ops.py", line 26, in <module> from tensorflow.python.data.experimental.ops import compression_ops File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\experimental\ops\compression_ops.py", line 20, in <module> from tensorflow.python.data.util import structure File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\util\structure.py", line 26, in <module> from tensorflow.python.data.util import nest File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\util\nest.py", line 40, in <module> from tensorflow.python.framework import sparse_tensor as _sparse_tensor File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\framework\sparse_tensor.py", line 28, in <module> from tensorflow.python.framework import constant_op File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\framework\constant_op.py", line 29, in <module> from tensorflow.python.eager import execute File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\eager\execute.py", line 27, in <module> from tensorflow.python.framework import dtypes File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\framework\dtypes.py", line 585, in <module> np.object, File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\numpy\__init__.py", line 324, in __getattr__ raise AttributeError(__former_attrs__[attr]) AttributeError: module 'numpy' has no attribute 'object'. `np.object` was a deprecated alias for the builtin `object`. To avoid this error in existing code, use `object` by itself. Doing this will not modify any behavior and is safe. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations ### Standalone code to reproduce the issue ```shell import tensorflow as tf and the version of numpy is 1.26. ``` ### Relevant log output ```shell import tensorflow as tf C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\framework\dtypes.py:585: FutureWarning: In the future `np.object` will be defined as the corresponding NumPy scalar. np.object, Traceback (most recent call last): File "<stdin>", line 1, in <module> File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\__init__.py", line 41, in <module> from tensorflow.python.tools import module_util as _module_util File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\__init__.py", line 46, in <module> from tensorflow.python import data File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\__init__.py", line 25, in <module> from tensorflow.python.data import experimental File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\experimental\__init__.py", line 97, in <module> from tensorflow.python.data.experimental import service File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\experimental\service\__init__.py", line 353, in <module> from tensorflow.python.data.experimental.ops.data_service_ops import distribute File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\experimental\ops\data_service_ops.py", line 26, in <module> from tensorflow.python.data.experimental.ops import compression_ops File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\experimental\ops\compression_ops.py", line 20, in <module> from tensorflow.python.data.util import structure File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\util\structure.py", line 26, in <module> from tensorflow.python.data.util import nest File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\data\util\nest.py", line 40, in <module> from tensorflow.python.framework import sparse_tensor as _sparse_tensor File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\framework\sparse_tensor.py", line 28, in <module> from tensorflow.python.framework import constant_op File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\framework\constant_op.py", line 29, in <module> from tensorflow.python.eager import execute File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\eager\execute.py", line 27, in <module> from tensorflow.python.framework import dtypes File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\tensorflow\python\framework\dtypes.py", line 585, in <module> np.object, File "C:\Users\dell\.conda\envs\tf-gpu\lib\site-packages\numpy\__init__.py", line 324, in __getattr__ raise AttributeError(__former_attrs__[attr]) AttributeError: module 'numpy' has no attribute 'object'. `np.object` was a deprecated alias for the builtin `object`. To avoid this error in existing code, use `object` by itself. Doing this will not modify any behavior and is safe. The aliases was originally deprecated in NumPy 1.20; for more details and guidance see the original release note at: https://numpy.org/devdocs/release/1.20.0-notes.html#deprecations ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62479/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62479/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62478	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62478/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62478/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62478/events	https://github.com/tensorflow/tensorflow/issues/62478	2,011,111,656	I_kwDOArmXAs533yDo	62,478	Inconsistency in Output of `tf.nn.conv2d + tf.cos` Combination in XLA Compiled Models on GPU	{ "login": "Gwihwan-Go", "id": 83144588, "node_id": "MDQ6VXNlcjgzMTQ0NTg4", "avatar_url": "https://avatars.githubusercontent.com/u/83144588?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Gwihwan-Go", "html_url": "https://github.com/Gwihwan-Go", "followers_url": "https://api.github.com/users/Gwihwan-Go/followers", "following_url": "https://api.github.com/users/Gwihwan-Go/following{/other_user}", "gists_url": "https://api.github.com/users/Gwihwan-Go/gists{/gist_id}", "starred_url": "https://api.github.com/users/Gwihwan-Go/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Gwihwan-Go/subscriptions", "organizations_url": "https://api.github.com/users/Gwihwan-Go/orgs", "repos_url": "https://api.github.com/users/Gwihwan-Go/repos", "events_url": "https://api.github.com/users/Gwihwan-Go/events{/privacy}", "received_events_url": "https://api.github.com/users/Gwihwan-Go/received_events", "type": "User", "site_admin": false }	[ { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1133285679, "node_id": "MDU6TGFiZWwxMTMzMjg1Njc5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:xla", "name": "comp:xla", "color": "0052cc", "default": false, "description": "XLA" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "We found another code that triggers same bug, which does not require specific input conditions. Below is the standalone code responsible for this issue:\r\n```python\r\nfrom typing import Dict\r\nimport tensorflow as tf\r\nimport pickle\r\nimport os\r\nimport numpy as np\r\n\r\nclass Model1(tf.keras.Model):\r\n def __init__(self):\r\n super().__init__()\r\n # Tensor objects (with comments for shapes)\r\n\r\n @tf.function(jit_compile=True)\r\n def __call__(self, inp1, inp2):\r\n # Forward pass logic using TensorFlow operations\r\n\r\n nnconv = tf.nn.conv2d(inp2, inp1, strides=1, padding=\"VALID\", dilations=(2, 1))\r\n _convcos = tf.cos(nnconv)\r\n return nnconv, _convcos\r\n\r\ninputs = [\r\ntf.random.uniform(shape=[2, 120, 120, 2], dtype=tf.float32),\r\ntf.random.uniform(shape=[1, 3, 140, 120], dtype=tf.float32),\r\n]\r\nmodel1 = Model1()\r\ndevice = \"gpu\"\r\nwith tf.device(device):\r\n tf.config.run_functions_eagerly(True)\r\n out1 = model1(inputs)\r\n out2 = model1(inputs)\r\n print(f'=========eager_output(version:{tf.__version__})================')\r\n try :\r\n for i in range(min(len(out1),len(out2))):\r\n np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th')\r\n print(\"XLA_eager does not trigger assertion\")\r\n except AssertionError as e:\r\n print(\"XLA_eager triggers assertion\")\r\n print(e)\r\n tf.config.run_functions_eagerly(False)\r\n out1 = model1(inputs)\r\n out2 = model1(inputs)\r\n print(f'=========compiled_output(version:{tf.__version__})================')\r\n try :\r\n for i in range(min(len(out1),len(out2))):\r\n np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th')\r\n print(\"XLA_complie does not trigger assertion\")\r\n except AssertionError as e:\r\n print(\"XLA_complie triggers assertion\")\r\n print(e)\r\n```", "@SuryanarayanaY Hello, Have you reproduced this issue?", "Hi @GwiHwan-Go ,\r\n\r\nI have tested with Tf2.14v and GPU runtime and the difference is related to precision only which happens due to XLA internal fusions and conversions. If I change atol to 0.01 from 0.001 no assertion error will be raises.\r\n\r\nAlso I have printed `tf.reduce_sum(out1).numpy()` and `tf.reduce_sum(out2).numpy()` which is printing `303656.34` and `303656.3` respectively. This indicates the results are same but only precision differences with XLA which is expected.\r\n\r\nPlease refer attached [gist](https://colab.sandbox.google.com/gist/SuryanarayanaY/b4854039d081d04d1c63ae112ff6de25/62478_r1-2-14.ipynb).\r\n\r\nCould you please check the same with TF2.15v also and confirm the results are only minor precision differences or not.\r\n\r\nThanks!", "Thank you for your response! As it is not an bug, I'm closing the issue.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62478\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62478\">No</a>\n" ]	2023-11-26T16:42:31	2023-12-06T12:50:37	2023-12-06T12:50:34	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.15.0 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04.3 LTS ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version cuda : 12.2 / cudnn 8.9.04 ### GPU model and memory Tesla V100S-PCIE-32GB ### Current behavior? I've identified a critical issue in TensorFlow 2.15.0 where the combination of `tf.nn.conv2d` and `tf.cos` in an XLA compiled model produces significantly different outputs compared to the eager execution mode. This inconsistency is particularly concerning given that the data type used is `float32`, which is a standard in many applications. This issue only occurs with certain input data with gpu device To reproduce this, please download first [pickle file](https://github.com/GwiHwan-Go/repo/raw/main/conv_cos_bug_inputs.pickle) and replace `YOUR_PICKLE_FILE_PATH` with your pickle file path. ### Standalone code to reproduce the issue ```shell import tensorflow as tf import pickle import os import numpy as np class Model1(tf.keras.Model): def __init__(self): super().__init__() @tf.function(jit_compile=True) def __call__(self, inp1, inp2): conv2 = tf.nn.conv2d(inp1, inp2, strides=1, padding="VALID", dilations=(1, 3)) _cos = tf.cos(conv2) return conv2, _cos model1 = Model1() device = "gpu" # Path to the pickle file relative to the script directory pickle_file_path = 'conv_cos_bug_inputs.pickle' # YOUR_PICKLE_FILE_PATH if not os.path.exists(pickle_file_path) : print(f'Pickle file not exist') else : with open(pickle_file_path, 'rb') as f : oracle = pickle.load(f) inputs = [tf.convert_to_tensor(arr) for arr in oracle] with tf.device(device): tf.config.run_functions_eagerly(True) out1 = model1(inputs) out2 = model1(inputs) print(f'=========eager_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th') print("XLA_eager does not trigger assertion") except AssertionError as e: print("XLA_eager triggers assertion") print(e) tf.config.run_functions_eagerly(False) out1 = model1(inputs) out2 = model1(inputs) print(f'=========compiled_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th') print("XLA_complie does not trigger assertion") except AssertionError as e: print("XLA_complie triggers assertion") print(e) ``` ### Relevant log output ```shell =========eager_output(version:2.15.0)================ XLA_eager does not trigger assertion 2023-11-26 16:40:44.610101: I external/local_xla/xla/service/service.cc:168] XLA service 0x559e2c9bb480 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: 2023-11-26 16:40:44.610131: I external/local_xla/xla/service/service.cc:176] StreamExecutor device (0): Tesla V100S-PCIE-32GB, Compute Capability 7.0 =========compiled_output(version:2.15.0)================ XLA_complie triggers assertion Not equal to tolerance rtol=0.001, atol=0.001 at checking 1th Mismatched elements: 4113 / 10000 (41.1%) Max absolute difference: 0.03115243 Max relative difference: 129.51477 x: array([[[[ 0.87046 , -0.997616, 0.527893, ..., -0.624568, -0.543016, -0.689372], [-0.174798, -0.671232, 0.72486 , ..., 0.273734, 0.056437,... y: array([[[[ 0.87142 , -0.997339, 0.534512, ..., -0.624568, -0.536439, -0.686538], [-0.186324, -0.671232, 0.727545, ..., 0.273734, 0.056437,... ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62478/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62478/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62477	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62477/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62477/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62477/events	https://github.com/tensorflow/tensorflow/issues/62477	2,011,099,875	I_kwDOArmXAs533vLj	62,477	Inconsistent Outputs from `tf.nn.conv2d_transpose` on GPU	{ "login": "Gwihwan-Go", "id": 83144588, "node_id": "MDQ6VXNlcjgzMTQ0NTg4", "avatar_url": "https://avatars.githubusercontent.com/u/83144588?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Gwihwan-Go", "html_url": "https://github.com/Gwihwan-Go", "followers_url": "https://api.github.com/users/Gwihwan-Go/followers", "following_url": "https://api.github.com/users/Gwihwan-Go/following{/other_user}", "gists_url": "https://api.github.com/users/Gwihwan-Go/gists{/gist_id}", "starred_url": "https://api.github.com/users/Gwihwan-Go/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Gwihwan-Go/subscriptions", "organizations_url": "https://api.github.com/users/Gwihwan-Go/orgs", "repos_url": "https://api.github.com/users/Gwihwan-Go/repos", "events_url": "https://api.github.com/users/Gwihwan-Go/events{/privacy}", "received_events_url": "https://api.github.com/users/Gwihwan-Go/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 1097547538, "node_id": "MDU6TGFiZWwxMDk3NTQ3NTM4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:gpu", "name": "comp:gpu", "color": "0052cc", "default": false, "description": "GPU related issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@GwiHwan-Go,\r\nI was able to replicate the reported behaviour with TF v2.13, TF v2.14 and tf-nightly..\r\nCurrently we are investigating the issue & will deep dive into the issue and provide the root-cause for the same. Thank you!", "> @GwiHwan-Go, I was able to replicate the reported behaviour with TF v2.13, TF v2.14 and tf-nightly.. Currently we are investigating the issue & will deep dive into the issue and provide the root-cause for the same. Thank you!\r\n\r\nThank you for your response! Please don't hesitate to request any additional information if needed." ]	2023-11-26T16:09:08	2023-12-05T23:14:19	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.15.0 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04.3 LTS ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version cuda : 12.2 / cudnn 8.9.04 ### GPU model and memory Tesla V100S-PCIE-32GB ### Current behavior? I've encountered an issue where `tf.nn.conv2d_transpose` produces inconsistent outputs on a GPU device across different runs, despite using the same input and operator. This inconsistency is seen both on xla compiled mode and eager mode. ### Standalone code to reproduce the issue ```shell import tensorflow as tf import numpy as np class Model1(tf.keras.Model): def __init__(self): super().__init__() @tf.function(jit_compile=True) def __call__(self, inp): testout = tf.nn.conv2d_transpose(inp, inp, output_shape=[26, 57, 32, 1], strides=1, padding="VALID") return testout inputs = [ tf.random.uniform(shape=[26, 32, 1, 9], dtype=tf.float16), ] model1 = Model1() device = "gpu" with tf.device(device): tf.config.run_functions_eagerly(True) out1 = model1(inputs) out2 = model1(inputs) print(f'=========eager_output(version:{tf.__version__})================') print(tf.sysconfig.get_build_info()["cuda_version"]) print(tf.sysconfig.get_build_info()["cudnn_version"]) try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th') print("XLA_eager does not trigger assertion") except AssertionError as e: print("XLA_eager triggers assertion") print(e) tf.config.run_functions_eagerly(False) out1 = model1(inputs) out2 = model1(inputs) print(f'=========compiled_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.001, atol=0.001, err_msg=f'at checking {i}th') print("XLA_complie does not trigger assertion") except AssertionError as e: print("XLA_complie triggers assertion") print(e) ``` ### Relevant log output ```shell =========eager_output(version:2.15.0)================ 12.2 8 XLA_eager triggers assertion Not equal to tolerance rtol=0.001, atol=0.001 at checking 0th Mismatched elements: 178 / 1824 (9.76%) Max absolute difference: 0.125 Max relative difference: 0.002499 x: array([[[3.156], [2.463], [1.905],... y: array([[[3.156], [2.463], [1.905],... 2023-11-26 16:06:46.480802: I external/local_xla/xla/service/service.cc:168] XLA service 0x557fce97e690 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: 2023-11-26 16:06:46.480861: I external/local_xla/xla/service/service.cc:176] StreamExecutor device (0): Tesla V100S-PCIE-32GB, Compute Capability 7.0 WARNING: All log messages before absl::InitializeLog() is called are written to STDERR I0000 00:00:1701014806.982871 61794 device_compiler.h:186] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process. =========compiled_output(version:2.15.0)================ XLA_complie triggers assertion Not equal to tolerance rtol=0.001, atol=0.001 at checking 0th Mismatched elements: 133 / 1824 (7.29%) Max absolute difference: 0.125 Max relative difference: 0.002611 x: array([[[3.156], [2.463], [1.905],... y: array([[[3.156], [2.463], [1.905],... ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62477/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62477/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62476	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62476/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62476/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62476/events	https://github.com/tensorflow/tensorflow/issues/62476	2,010,451,473	I_kwDOArmXAs531Q4R	62,476	ValueError: Name tf.RaggedTensorSpec has already been registered for class tensorflow.python.ops.ragged.ragged_tensor.RaggedTensorSpec.	{ "login": "JT-Studios", "id": 106934268, "node_id": "U_kgDOBl-v_A", "avatar_url": "https://avatars.githubusercontent.com/u/106934268?v=4", "gravatar_id": "", "url": "https://api.github.com/users/JT-Studios", "html_url": "https://github.com/JT-Studios", "followers_url": "https://api.github.com/users/JT-Studios/followers", "following_url": "https://api.github.com/users/JT-Studios/following{/other_user}", "gists_url": "https://api.github.com/users/JT-Studios/gists{/gist_id}", "starred_url": "https://api.github.com/users/JT-Studios/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/JT-Studios/subscriptions", "organizations_url": "https://api.github.com/users/JT-Studios/orgs", "repos_url": "https://api.github.com/users/JT-Studios/repos", "events_url": "https://api.github.com/users/JT-Studios/events{/privacy}", "received_events_url": "https://api.github.com/users/JT-Studios/received_events", "type": "User", "site_admin": false }	[ { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 5508003926, "node_id": "LA_kwDOArmXAs8AAAABSE14Vg", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.13", "name": "TF 2.13", "color": "B13ACB", "default": false, "description": "For issues related to Tensorflow 2.13" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @JT-Studios ,\r\n\r\nFrom the error log ,the error is generating from importing tensorflow. This might be due to different compatible version of keras vs tensorflow. Could you please try uninstalling tensorflow and keras packages and install tensorflow again and try importing tensorflow again and let us know if still issue persists.\r\n\r\nThanks!", "I have reinstalled tensorflow however when I try to use the import commands they won't stop running. After a minute I interrupted the kernel and it gave me this output:\r\n\r\n`---------------------------------------------------------------------------\r\nKeyboardInterrupt Traceback (most recent call last)\r\nCell In[5], line 5\r\n 1 #!pwd\r\n 2 #print (PYTHONPATH)\r\n 3 #print(os.path)\r\n----> 5 import tensorflow as tf\r\n 6 from object_detection.utils import config_util\r\n 7 from object_detection.protos import pipeline_pb2\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/__init__.py:54\r\n 52 from tensorflow._api.v2 import autograph\r\n 53 from tensorflow._api.v2 import bitwise\r\n---> 54 from tensorflow._api.v2 import compat\r\n 55 from tensorflow._api.v2 import config\r\n 56 from tensorflow._api.v2 import data\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/compat/__init__.py:8\r\n 3 \"\"\"Public API for tf._api.v2.compat namespace\r\n 4 \"\"\"\r\n 6 import sys as _sys\r\n----> 8 from tensorflow._api.v2.compat import v1\r\n 9 from tensorflow._api.v2.compat import v2\r\n 10 from tensorflow.python.compat.compat import forward_compatibility_horizon # line: 125\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/compat/v1/__init__.py:32\r\n 30 from tensorflow._api.v2.compat.v1 import autograph\r\n 31 from tensorflow._api.v2.compat.v1 import bitwise\r\n---> 32 from tensorflow._api.v2.compat.v1 import compat\r\n 33 from tensorflow._api.v2.compat.v1 import config\r\n 34 from tensorflow._api.v2.compat.v1 import data\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/compat/v1/compat/__init__.py:9\r\n 6 import sys as _sys\r\n 8 from tensorflow._api.v2.compat.v1.compat import v1\r\n----> 9 from tensorflow._api.v2.compat.v1.compat import v2\r\n 10 from tensorflow.python.compat.compat import forward_compatibility_horizon # line: 125\r\n 11 from tensorflow.python.compat.compat import forward_compatible # line: 65\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/compat/v1/compat/v2/__init__.py:28\r\n 25 from tensorflow.python.util.lazy_loader import LazyLoader as _LazyLoader\r\n 26 from tensorflow.python.util.lazy_loader import KerasLazyLoader as _KerasLazyLoader\r\n---> 28 from tensorflow._api.v2.compat.v2 import __internal__\r\n 29 from tensorflow._api.v2.compat.v2 import __operators__\r\n 30 from tensorflow._api.v2.compat.v2 import audio\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/compat/v2/__init__.py:34\r\n 32 from tensorflow._api.v2.compat.v2 import autograph\r\n 33 from tensorflow._api.v2.compat.v2 import bitwise\r\n---> 34 from tensorflow._api.v2.compat.v2 import compat\r\n 35 from tensorflow._api.v2.compat.v2 import config\r\n 36 from tensorflow._api.v2.compat.v2 import data\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/compat/v2/compat/__init__.py:9\r\n 6 import sys as _sys\r\n 8 from tensorflow._api.v2.compat.v2.compat import v1\r\n----> 9 from tensorflow._api.v2.compat.v2.compat import v2\r\n 10 from tensorflow.python.compat.compat import forward_compatibility_horizon # line: 125\r\n 11 from tensorflow.python.compat.compat import forward_compatible # line: 65\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/compat/v2/compat/v2/__init__.py:65\r\n 63 from tensorflow._api.v2.compat.v2 import strings\r\n 64 from tensorflow._api.v2.compat.v2 import summary\r\n---> 65 from tensorflow._api.v2.compat.v2 import sysconfig\r\n 66 from tensorflow._api.v2.compat.v2 import test\r\n 67 from tensorflow._api.v2.compat.v2 import tpu\r\n\r\nFile <frozen importlib._bootstrap>:1176, in _find_and_load(name, import_)\r\n\r\nFile <frozen importlib._bootstrap>:1147, in _find_and_load_unlocked(name, import_)\r\n\r\nFile <frozen importlib._bootstrap>:690, in _load_unlocked(spec)\r\n\r\nFile <frozen importlib._bootstrap_external>:936, in exec_module(self, module)\r\n\r\nFile <frozen importlib._bootstrap_external>:1032, in get_code(self, fullname)\r\n\r\nFile <frozen importlib._bootstrap_external>:1131, in get_data(self, path)\r\n\r\nKeyboardInterrupt: \r\n`\r\n\r\nIt is interesting to note that it is running 3.11 files even though this is not the version it is supposed to be working in.", "Update: I have created a new virtual environment and the issue is resolved but a new issue started appearing:\r\n\r\n`---------------------------------------------------------------------------\r\nTypeError Traceback (most recent call last)\r\nCell In[29], line 2\r\n 1 import tensorflow as tf\r\n----> 2 from object_detection.utils import config_util\r\n 3 from object_detection.protos import pipeline_pb2\r\n 4 from google.protobuf import text_format\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/object_detection/utils/config_util.py:24\r\n 20 from google.protobuf import text_format\r\n 22 from tensorflow.python.lib.io import file_io\r\n---> 24 from object_detection.protos import eval_pb2\r\n 25 from object_detection.protos import graph_rewriter_pb2\r\n 26 from object_detection.protos import input_reader_pb2\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/object_detection/protos/eval_pb2.py:36\r\n 13 _sym_db = _symbol_database.Default()\r\n 18 DESCRIPTOR = _descriptor.FileDescriptor(\r\n 19 name='object_detection/protos/eval.proto',\r\n 20 package='object_detection.protos',\r\n (...)\r\n 23 serialized_pb=_b('\\n\\\"object_detection/protos/eval.proto\\x12\\x17object_detection.protos\\\"\\xf7\\x05\\n\\nEvalConfig\\x12\\x15\\n\\nbatch_size\\x18\\x19 \\x01(\\r:\\x01\\x31\\x12\\x1e\\n\\x12num_visualizations\\x18\\x01 \\x01(\\r:\\x02\\x31\\x30\\x12\\x1e\\n\\x0cnum_examples\\x18\\x02 \\x01(\\r:\\x04\\x35\\x30\\x30\\x30\\x42\\x02\\x18\\x01\\x12\\x1f\\n\\x12\\x65val_interval_secs\\x18\\x03 \\x01(\\r:\\x03\\x33\\x30\\x30\\x12\\x18\\n\\tmax_evals\\x18\\x04 \\x01(\\r:\\x01\\x30\\x42\\x02\\x18\\x01\\x12\\x19\\n\\nsave_graph\\x18\\x05 \\x01(\\x08:\\x05\\x66\\x61lse\\x12\\\"\\n\\x18visualization_export_dir\\x18\\x06 \\x01(\\t:\\x00\\x12\\x15\\n\\x0b\\x65val_master\\x18\\x07 \\x01(\\t:\\x00\\x12\\x13\\n\\x0bmetrics_set\\x18\\x08 \\x03(\\t\\x12\\x15\\n\\x0b\\x65xport_path\\x18\\t \\x01(\\t:\\x00\\x12!\\n\\x12ignore_groundtruth\\x18\\n \\x01(\\x08:\\x05\\x66\\x61lse\\x12\\\"\\n\\x13use_moving_averages\\x18\\x0b \\x01(\\x08:\\x05\\x66\\x61lse\\x12\\\"\\n\\x13\\x65val_instance_masks\\x18\\x0c \\x01(\\x08:\\x05\\x66\\x61lse\\x12 \\n\\x13min_score_threshold\\x18\\r \\x01(\\x02:\\x03\\x30.5\\x12&\\n\\x1amax_num_boxes_to_visualize\\x18\\x0e \\x01(\\x05:\\x02\\x32\\x30\\x12\\x1a\\n\\x0bskip_scores\\x18\\x0f \\x01(\\x08:\\x05\\x66\\x61lse\\x12\\x1a\\n\\x0bskip_labels\\x18\\x10 \\x01(\\x08:\\x05\\x66\\x61lse\\x12*\\n\\x1bvisualize_groundtruth_boxes\\x18\\x11 \\x01(\\x08:\\x05\\x66\\x61lse\\x12\\x32\\n#groundtruth_box_visualization_color\\x18\\x12 \\x01(\\t:\\x05\\x62lack\\x12\\x35\\n&keep_image_id_for_visualization_export\\x18\\x13 \\x01(\\x08:\\x05\\x66\\x61lse\\x12$\\n\\x16retain_original_images\\x18\\x17 \\x01(\\x08:\\x04true\\x12+\\n\\x1cinclude_metrics_per_category\\x18\\x18 \\x01(\\x08:\\x05\\x66\\x61lse')\r\n 24 )\r\n 29 _EVALCONFIG = _descriptor.Descriptor(\r\n 30 name='EvalConfig',\r\n 31 full_name='object_detection.protos.EvalConfig',\r\n 32 filename=None,\r\n 33 file=DESCRIPTOR,\r\n 34 containing_type=None,\r\n 35 fields=[\r\n---> 36 _descriptor.FieldDescriptor(\r\n 37 name='batch_size', full_name='object_detection.protos.EvalConfig.batch_size', index=0,\r\n 38 number=25, type=13, cpp_type=3, label=1,\r\n 39 has_default_value=True, default_value=1,\r\n 40 message_type=None, enum_type=None, containing_type=None,\r\n 41 is_extension=False, extension_scope=None,\r\n 42 serialized_options=None, file=DESCRIPTOR),\r\n 43 _descriptor.FieldDescriptor(\r\n 44 name='num_visualizations', full_name='object_detection.protos.EvalConfig.num_visualizations', index=1,\r\n 45 number=1, type=13, cpp_type=3, label=1,\r\n 46 has_default_value=True, default_value=10,\r\n 47 message_type=None, enum_type=None, containing_type=None,\r\n 48 is_extension=False, extension_scope=None,\r\n 49 serialized_options=None, file=DESCRIPTOR),\r\n 50 _descriptor.FieldDescriptor(\r\n 51 name='num_examples', full_name='object_detection.protos.EvalConfig.num_examples', index=2,\r\n 52 number=2, type=13, cpp_type=3, label=1,\r\n 53 has_default_value=True, default_value=5000,\r\n 54 message_type=None, enum_type=None, containing_type=None,\r\n 55 is_extension=False, extension_scope=None,\r\n 56 serialized_options=_b('\\030\\001'), file=DESCRIPTOR),\r\n 57 _descriptor.FieldDescriptor(\r\n 58 name='eval_interval_secs', full_name='object_detection.protos.EvalConfig.eval_interval_secs', index=3,\r\n 59 number=3, type=13, cpp_type=3, label=1,\r\n 60 has_default_value=True, default_value=300,\r\n 61 message_type=None, enum_type=None, containing_type=None,\r\n 62 is_extension=False, extension_scope=None,\r\n 63 serialized_options=None, file=DESCRIPTOR),\r\n 64 _descriptor.FieldDescriptor(\r\n 65 name='max_evals', full_name='object_detection.protos.EvalConfig.max_evals', index=4,\r\n 66 number=4, type=13, cpp_type=3, label=1,\r\n 67 has_default_value=True, default_value=0,\r\n 68 message_type=None, enum_type=None, containing_type=None,\r\n 69 is_extension=False, extension_scope=None,\r\n 70 serialized_options=_b('\\030\\001'), file=DESCRIPTOR),\r\n 71 _descriptor.FieldDescriptor(\r\n 72 name='save_graph', full_name='object_detection.protos.EvalConfig.save_graph', index=5,\r\n 73 number=5, type=8, cpp_type=7, label=1,\r\n 74 has_default_value=True, default_value=False,\r\n 75 message_type=None, enum_type=None, containing_type=None,\r\n 76 is_extension=False, extension_scope=None,\r\n 77 serialized_options=None, file=DESCRIPTOR),\r\n 78 _descriptor.FieldDescriptor(\r\n 79 name='visualization_export_dir', full_name='object_detection.protos.EvalConfig.visualization_export_dir', index=6,\r\n 80 number=6, type=9, cpp_type=9, label=1,\r\n 81 has_default_value=True, default_value=_b(\"\").decode('utf-8'),\r\n 82 message_type=None, enum_type=None, containing_type=None,\r\n 83 is_extension=False, extension_scope=None,\r\n 84 serialized_options=None, file=DESCRIPTOR),\r\n 85 _descriptor.FieldDescriptor(\r\n 86 name='eval_master', full_name='object_detection.protos.EvalConfig.eval_master', index=7,\r\n 87 number=7, type=9, cpp_type=9, label=1,\r\n 88 has_default_value=True, default_value=_b(\"\").decode('utf-8'),\r\n 89 message_type=None, enum_type=None, containing_type=None,\r\n 90 is_extension=False, extension_scope=None,\r\n 91 serialized_options=None, file=DESCRIPTOR),\r\n 92 _descriptor.FieldDescriptor(\r\n 93 name='metrics_set', full_name='object_detection.protos.EvalConfig.metrics_set', index=8,\r\n 94 number=8, type=9, cpp_type=9, label=3,\r\n 95 has_default_value=False, default_value=[],\r\n 96 message_type=None, enum_type=None, containing_type=None,\r\n 97 is_extension=False, extension_scope=None,\r\n 98 serialized_options=None, file=DESCRIPTOR),\r\n 99 _descriptor.FieldDescriptor(\r\n 100 name='export_path', full_name='object_detection.protos.EvalConfig.export_path', index=9,\r\n 101 number=9, type=9, cpp_type=9, label=1,\r\n 102 has_default_value=True, default_value=_b(\"\").decode('utf-8'),\r\n 103 message_type=None, enum_type=None, containing_type=None,\r\n 104 is_extension=False, extension_scope=None,\r\n 105 serialized_options=None, file=DESCRIPTOR),\r\n 106 _descriptor.FieldDescriptor(\r\n 107 name='ignore_groundtruth', full_name='object_detection.protos.EvalConfig.ignore_groundtruth', index=10,\r\n 108 number=10, type=8, cpp_type=7, label=1,\r\n 109 has_default_value=True, default_value=False,\r\n 110 message_type=None, enum_type=None, containing_type=None,\r\n 111 is_extension=False, extension_scope=None,\r\n 112 serialized_options=None, file=DESCRIPTOR),\r\n 113 _descriptor.FieldDescriptor(\r\n 114 name='use_moving_averages', full_name='object_detection.protos.EvalConfig.use_moving_averages', index=11,\r\n 115 number=11, type=8, cpp_type=7, label=1,\r\n 116 has_default_value=True, default_value=False,\r\n 117 message_type=None, enum_type=None, containing_type=None,\r\n 118 is_extension=False, extension_scope=None,\r\n 119 serialized_options=None, file=DESCRIPTOR),\r\n 120 _descriptor.FieldDescriptor(\r\n 121 name='eval_instance_masks', full_name='object_detection.protos.EvalConfig.eval_instance_masks', index=12,\r\n 122 number=12, type=8, cpp_type=7, label=1,\r\n 123 has_default_value=True, default_value=False,\r\n 124 message_type=None, enum_type=None, containing_type=None,\r\n 125 is_extension=False, extension_scope=None,\r\n 126 serialized_options=None, file=DESCRIPTOR),\r\n 127 _descriptor.FieldDescriptor(\r\n 128 name='min_score_threshold', full_name='object_detection.protos.EvalConfig.min_score_threshold', index=13,\r\n 129 number=13, type=2, cpp_type=6, label=1,\r\n 130 has_default_value=True, default_value=float(0.5),\r\n 131 message_type=None, enum_type=None, containing_type=None,\r\n 132 is_extension=False, extension_scope=None,\r\n 133 serialized_options=None, file=DESCRIPTOR),\r\n 134 _descriptor.FieldDescriptor(\r\n 135 name='max_num_boxes_to_visualize', full_name='object_detection.protos.EvalConfig.max_num_boxes_to_visualize', index=14,\r\n 136 number=14, type=5, cpp_type=1, label=1,\r\n 137 has_default_value=True, default_value=20,\r\n 138 message_type=None, enum_type=None, containing_type=None,\r\n 139 is_extension=False, extension_scope=None,\r\n 140 serialized_options=None, file=DESCRIPTOR),\r\n 141 _descriptor.FieldDescriptor(\r\n 142 name='skip_scores', full_name='object_detection.protos.EvalConfig.skip_scores', index=15,\r\n 143 number=15, type=8, cpp_type=7, label=1,\r\n 144 has_default_value=True, default_value=False,\r\n 145 message_type=None, enum_type=None, containing_type=None,\r\n 146 is_extension=False, extension_scope=None,\r\n 147 serialized_options=None, file=DESCRIPTOR),\r\n 148 _descriptor.FieldDescriptor(\r\n 149 name='skip_labels', full_name='object_detection.protos.EvalConfig.skip_labels', index=16,\r\n 150 number=16, type=8, cpp_type=7, label=1,\r\n 151 has_default_value=True, default_value=False,\r\n 152 message_type=None, enum_type=None, containing_type=None,\r\n 153 is_extension=False, extension_scope=None,\r\n 154 serialized_options=None, file=DESCRIPTOR),\r\n 155 _descriptor.FieldDescriptor(\r\n 156 name='visualize_groundtruth_boxes', full_name='object_detection.protos.EvalConfig.visualize_groundtruth_boxes', index=17,\r\n 157 number=17, type=8, cpp_type=7, label=1,\r\n 158 has_default_value=True, default_value=False,\r\n 159 message_type=None, enum_type=None, containing_type=None,\r\n 160 is_extension=False, extension_scope=None,\r\n 161 serialized_options=None, file=DESCRIPTOR),\r\n 162 _descriptor.FieldDescriptor(\r\n 163 name='groundtruth_box_visualization_color', full_name='object_detection.protos.EvalConfig.groundtruth_box_visualization_color', index=18,\r\n 164 number=18, type=9, cpp_type=9, label=1,\r\n 165 has_default_value=True, default_value=_b(\"black\").decode('utf-8'),\r\n 166 message_type=None, enum_type=None, containing_type=None,\r\n 167 is_extension=False, extension_scope=None,\r\n 168 serialized_options=None, file=DESCRIPTOR),\r\n 169 _descriptor.FieldDescriptor(\r\n 170 name='keep_image_id_for_visualization_export', full_name='object_detection.protos.EvalConfig.keep_image_id_for_visualization_export', index=19,\r\n 171 number=19, type=8, cpp_type=7, label=1,\r\n 172 has_default_value=True, default_value=False,\r\n 173 message_type=None, enum_type=None, containing_type=None,\r\n 174 is_extension=False, extension_scope=None,\r\n 175 serialized_options=None, file=DESCRIPTOR),\r\n 176 _descriptor.FieldDescriptor(\r\n 177 name='retain_original_images', full_name='object_detection.protos.EvalConfig.retain_original_images', index=20,\r\n 178 number=23, type=8, cpp_type=7, label=1,\r\n 179 has_default_value=True, default_value=True,\r\n 180 message_type=None, enum_type=None, containing_type=None,\r\n 181 is_extension=False, extension_scope=None,\r\n 182 serialized_options=None, file=DESCRIPTOR),\r\n 183 _descriptor.FieldDescriptor(\r\n 184 name='include_metrics_per_category', full_name='object_detection.protos.EvalConfig.include_metrics_per_category', index=21,\r\n 185 number=24, type=8, cpp_type=7, label=1,\r\n 186 has_default_value=True, default_value=False,\r\n 187 message_type=None, enum_type=None, containing_type=None,\r\n 188 is_extension=False, extension_scope=None,\r\n 189 serialized_options=None, file=DESCRIPTOR),\r\n 190 ],\r\n 191 extensions=[\r\n 192 ],\r\n 193 nested_types=[],\r\n 194 enum_types=[\r\n 195 ],\r\n 196 serialized_options=None,\r\n 197 is_extendable=False,\r\n 198 syntax='proto2',\r\n 199 extension_ranges=[],\r\n 200 oneofs=[\r\n 201 ],\r\n 202 serialized_start=64,\r\n 203 serialized_end=823,\r\n 204 )\r\n 206 DESCRIPTOR.message_types_by_name['EvalConfig'] = _EVALCONFIG\r\n 207 _sym_db.RegisterFileDescriptor(DESCRIPTOR)\r\n\r\nFile ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/google/protobuf/descriptor.py:561, in FieldDescriptor.__new__(cls, name, full_name, index, number, type, cpp_type, label, default_value, message_type, enum_type, containing_type, is_extension, extension_scope, options, serialized_options, has_default_value, containing_oneof, json_name, file, create_key)\r\n 555 def __new__(cls, name, full_name, index, number, type, cpp_type, label,\r\n 556 default_value, message_type, enum_type, containing_type,\r\n 557 is_extension, extension_scope, options=None,\r\n 558 serialized_options=None,\r\n 559 has_default_value=True, containing_oneof=None, json_name=None,\r\n 560 file=None, create_key=None): # pylint: disable=redefined-builtin\r\n--> 561 _message.Message._CheckCalledFromGeneratedFile()\r\n 562 if is_extension:\r\n 563 return _message.default_pool.FindExtensionByName(full_name)\r\n\r\nTypeError: Descriptors cannot not be created directly.\r\nIf this call came from a _pb2.py file, your generated code is out of date and must be regenerated with protoc >= 3.19.0.\r\nIf you cannot immediately regenerate your protos, some other possible workarounds are:\r\n 1. Downgrade the protobuf package to 3.20.x or lower.\r\n 2. Set PROTOCOL_BUFFERS_PYTHON_IMPLEMENTATION=python (but this will use pure-Python parsing and will be much slower).\r\n\r\nMore information: https://developers.google.com/protocol-buffers/docs/news/2022-05-06#python-updates\r\n`\r\n\r\nSorry the error is a bit lengthy. How do I go about fixing this problem?", "Issue has been resolved thanks for the help!", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62476\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62476\">No</a>\n", "> Issue has been resolved thanks for the help!\r\n\r\nHey, How do you solved this error?? I am also getting the same error\r\n![Uploading n.png…]()\r\n" ]	2023-11-25T03:16:11	2024-02-21T11:45:35	2023-11-28T01:37:42	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version 2.13.0rc0 ### Custom code Yes ### OS platform and distribution MacOS ### Mobile device _No response_ ### Python version Python 3.9.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? This code should import tensor flow ### Standalone code to reproduce the issue ```shell import tensorflow as tf from object_detection.utils import config_util from object_detection.protos import pipeline_pb2 from google.protobuf import text_format ``` ### Relevant log output ```shell -------------------------------------------------------------------------- ValueError Traceback (most recent call last) Cell In[24], line 1 ----> 1 import tensorflow as tf 2 from object_detection.utils import config_util 3 from object_detection.protos import pipeline_pb2 File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/__init__.py:48 45 from tensorflow.python import tf2 as _tf2 46 _tf2.enable() ---> 48 from tensorflow._api.v2 import __internal__ 49 from tensorflow._api.v2 import __operators__ 50 from tensorflow._api.v2 import audio File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/__internal__/__init__.py:11 9 from tensorflow._api.v2.__internal__ import decorator 10 from tensorflow._api.v2.__internal__ import dispatch ---> 11 from tensorflow._api.v2.__internal__ import distribute 12 from tensorflow._api.v2.__internal__ import eager_context 13 from tensorflow._api.v2.__internal__ import feature_column File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/__internal__/distribute/__init__.py:8 3 """Public API for tf._api.v2.__internal__.distribute namespace 4 """ 6 import sys as _sys ----> 8 from tensorflow._api.v2.__internal__.distribute import combinations 9 from tensorflow._api.v2.__internal__.distribute import interim 10 from tensorflow._api.v2.__internal__.distribute import multi_process_runner File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/_api/v2/__internal__/distribute/combinations/__init__.py:8 3 """Public API for tf._api.v2.__internal__.distribute.combinations namespace 4 """ 6 import sys as _sys ----> 8 from tensorflow.python.distribute.combinations import env # line: 456 9 from tensorflow.python.distribute.combinations import generate # line: 365 10 from tensorflow.python.distribute.combinations import in_main_process # line: 418 File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/distribute/combinations.py:33 29 import six 32 from tensorflow.python.client import session ---> 33 from tensorflow.python.distribute import collective_all_reduce_strategy 34 from tensorflow.python.distribute import distribute_lib 35 from tensorflow.python.distribute import multi_process_runner File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/distribute/collective_all_reduce_strategy.py:25 23 from tensorflow.core.protobuf import tensorflow_server_pb2 24 from tensorflow.python.distribute import collective_util ---> 25 from tensorflow.python.distribute import cross_device_ops as cross_device_ops_lib 26 from tensorflow.python.distribute import cross_device_utils 27 from tensorflow.python.distribute import device_util File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/distribute/cross_device_ops.py:28 26 from tensorflow.python.client import device_lib 27 from tensorflow.python.distribute import collective_util ---> 28 from tensorflow.python.distribute import cross_device_utils 29 from tensorflow.python.distribute import device_util 30 from tensorflow.python.distribute import distribute_utils File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/distribute/cross_device_utils.py:22 19 from typing import Callable, List, Optional, Union 21 from tensorflow.python.distribute import collective_util ---> 22 from tensorflow.python.distribute import values as value_lib 23 from tensorflow.python.eager import backprop_util 24 from tensorflow.python.eager import context File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/distribute/values.py:23 21 from tensorflow.core.protobuf import struct_pb2 22 from tensorflow.python.distribute import device_util ---> 23 from tensorflow.python.distribute import distribute_lib 24 from tensorflow.python.distribute import packed_distributed_variable as packed 25 from tensorflow.python.distribute import reduce_util File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/distribute/distribute_lib.py:206 204 from tensorflow.python.autograph.core import ag_ctx as autograph_ctx 205 from tensorflow.python.autograph.impl import api as autograph --> 206 from tensorflow.python.data.ops import dataset_ops 207 from tensorflow.python.distribute import collective_util 208 from tensorflow.python.distribute import device_util File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/data/__init__.py:21 15 """`tf.data.Dataset` API for input pipelines. 16 17 See [Importing Data](https://tensorflow.org/guide/data) for an overview. 18 """ 20 # pylint: disable=unused-import ---> 21 from tensorflow.python.data import experimental 22 from tensorflow.python.data.ops.dataset_ops import AUTOTUNE 23 from tensorflow.python.data.ops.dataset_ops import Dataset File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/data/experimental/__init__.py:98 15 """Experimental API for building input pipelines. 16 17 This module contains experimental `Dataset` sources and transformations that can (...) 94 @@UNKNOWN_CARDINALITY 95 """ 97 # pylint: disable=unused-import ---> 98 from tensorflow.python.data.experimental import service 99 from tensorflow.python.data.experimental.ops.batching import dense_to_ragged_batch 100 from tensorflow.python.data.experimental.ops.batching import dense_to_sparse_batch File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/data/experimental/service/__init__.py:419 1 # Copyright 2020 The TensorFlow Authors. All Rights Reserved. 2 # 3 # Licensed under the Apache License, Version 2.0 (the "License"); (...) 13 # limitations under the License. 14 # ============================================================================== 15 """API for using the tf.data service. 16 17 This module contains: (...) 416 job of ParameterServerStrategy). 417 """ --> 419 from tensorflow.python.data.experimental.ops.data_service_ops import distribute 420 from tensorflow.python.data.experimental.ops.data_service_ops import from_dataset_id 421 from tensorflow.python.data.experimental.ops.data_service_ops import register_dataset File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/data/experimental/ops/data_service_ops.py:22 20 from tensorflow.core.protobuf import data_service_pb2 21 from tensorflow.python import tf2 ---> 22 from tensorflow.python.data.experimental.ops import compression_ops 23 from tensorflow.python.data.experimental.service import _pywrap_server_lib 24 from tensorflow.python.data.experimental.service import _pywrap_utils File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/data/experimental/ops/compression_ops.py:16 1 # Copyright 2020 The TensorFlow Authors. All Rights Reserved. 2 # 3 # Licensed under the Apache License, Version 2.0 (the "License"); (...) 13 # limitations under the License. 14 # ============================================================================== 15 """Ops for compressing and uncompressing dataset elements.""" ---> 16 from tensorflow.python.data.util import structure 17 from tensorflow.python.ops import gen_experimental_dataset_ops as ged_ops 20 def compress(element): File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/data/util/structure.py:32 30 from tensorflow.python.ops import resource_variable_ops 31 from tensorflow.python.ops import tensor_array_ops ---> 32 from tensorflow.python.ops.ragged import ragged_tensor 33 from tensorflow.python.platform import tf_logging as logging 34 from tensorflow.python.types import internal File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/ops/ragged/ragged_tensor.py:2320 2313 return tensors 2316 # =============================================================================== 2317 # RaggedTensorSpec 2318 # =============================================================================== 2319 @tf_export("RaggedTensorSpec") -> 2320 @type_spec_registry.register("tf.RaggedTensorSpec") 2321 class RaggedTensorSpec( 2322 type_spec.BatchableTypeSpec, internal_types.RaggedTensorSpec): 2323 """Type specification for a `tf.RaggedTensor`.""" 2325 __slots__ = [ 2326 "_shape", "_dtype", "_ragged_rank", "_row_splits_dtype", 2327 "_flat_values_spec" 2328 ] File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/tensorflow/python/framework/type_spec_registry.py:59, in register.<locals>.decorator_fn(cls) 56 raise ValueError("Class %s.%s has already been registered with name %s." % 57 (cls.__module__, cls.__name__, _TYPE_SPEC_TO_NAME[cls])) 58 if name in _NAME_TO_TYPE_SPEC: ---> 59 raise ValueError("Name %s has already been registered for class %s.%s." % 60 (name, _NAME_TO_TYPE_SPEC[name].__module__, 61 _NAME_TO_TYPE_SPEC[name].__name__)) 62 _TYPE_SPEC_TO_NAME[cls] = name 63 _NAME_TO_TYPE_SPEC[name] = cls ValueError: Name tf.RaggedTensorSpec has already been registered for class tensorflow.python.ops.ragged.ragged_tensor.RaggedTensorSpec. ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62476/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62476/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62475	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62475/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62475/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62475/events	https://github.com/tensorflow/tensorflow/issues/62475	2,009,146,247	I_kwDOArmXAs53wSOH	62,475	Discrepancies in Output due to Altered Multiplication Sequence in TensorFlow Operations	{ "login": "Gwihwan-Go", "id": 83144588, "node_id": "MDQ6VXNlcjgzMTQ0NTg4", "avatar_url": "https://avatars.githubusercontent.com/u/83144588?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Gwihwan-Go", "html_url": "https://github.com/Gwihwan-Go", "followers_url": "https://api.github.com/users/Gwihwan-Go/followers", "following_url": "https://api.github.com/users/Gwihwan-Go/following{/other_user}", "gists_url": "https://api.github.com/users/Gwihwan-Go/gists{/gist_id}", "starred_url": "https://api.github.com/users/Gwihwan-Go/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Gwihwan-Go/subscriptions", "organizations_url": "https://api.github.com/users/Gwihwan-Go/orgs", "repos_url": "https://api.github.com/users/Gwihwan-Go/repos", "events_url": "https://api.github.com/users/Gwihwan-Go/events{/privacy}", "received_events_url": "https://api.github.com/users/Gwihwan-Go/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 1133285679, "node_id": "MDU6TGFiZWwxMTMzMjg1Njc5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:xla", "name": "comp:xla", "color": "0052cc", "default": false, "description": "XLA" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	closed	false	{ "login": "cantonios", "id": 2538739, "node_id": "MDQ6VXNlcjI1Mzg3Mzk=", "avatar_url": "https://avatars.githubusercontent.com/u/2538739?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cantonios", "html_url": "https://github.com/cantonios", "followers_url": "https://api.github.com/users/cantonios/followers", "following_url": "https://api.github.com/users/cantonios/following{/other_user}", "gists_url": "https://api.github.com/users/cantonios/gists{/gist_id}", "starred_url": "https://api.github.com/users/cantonios/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cantonios/subscriptions", "organizations_url": "https://api.github.com/users/cantonios/orgs", "repos_url": "https://api.github.com/users/cantonios/repos", "events_url": "https://api.github.com/users/cantonios/events{/privacy}", "received_events_url": "https://api.github.com/users/cantonios/received_events", "type": "User", "site_admin": false }	[ { "login": "cantonios", "id": 2538739, "node_id": "MDQ6VXNlcjI1Mzg3Mzk=", "avatar_url": "https://avatars.githubusercontent.com/u/2538739?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cantonios", "html_url": "https://github.com/cantonios", "followers_url": "https://api.github.com/users/cantonios/followers", "following_url": "https://api.github.com/users/cantonios/following{/other_user}", "gists_url": "https://api.github.com/users/cantonios/gists{/gist_id}", "starred_url": "https://api.github.com/users/cantonios/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cantonios/subscriptions", "organizations_url": "https://api.github.com/users/cantonios/orgs", "repos_url": "https://api.github.com/users/cantonios/repos", "events_url": "https://api.github.com/users/cantonios/events{/privacy}", "received_events_url": "https://api.github.com/users/cantonios/received_events", "type": "User", "site_admin": false }, { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@sachinprasadhs I was able to replicate this issue on colab, please find the gist [here](https://colab.research.google.com/gist/sushreebarsa/c25fcb38679e26e98601fd94a22cffce/62475.ipynb). Thank you!", "Hi, \r\n\r\nThis issue is not specific to XLA, mismatch of result happens even without XLA, this is due to the multiple tf.multiply operations happening in the code of below two equations.\r\n\r\n```\r\nout1 = tf.multiply(tf.multiply(p0, inputs), tf.multiply(p0, p0))\r\n\r\nout2 = tf.multiply(p0, tf.multiply(inputs, tf.multiply(p0, p0)))\r\n```\r\n\r\nHere in both cases, the order of multiplication is performed first on the inner most `tf.multiply`, if two `tf.multiply` are in the same level, leftmost operation is performed first and then the value will be ready for the next operation.\r\nDuring this process, since both takes different code path and different level of multiply operations, there will be precision loss, which is expected and the difference which is observed in not too large.\r\n\r\nBelow is the simplified code which produces the similar behavior.\r\n\r\n```python\r\nimport tensorflow as tf\r\nimport numpy as np\r\nparams = [\r\ntf.random.uniform(shape=[10, 10], dtype=tf.float16),\r\n]\r\ninputs = [tf.constant(-0.7305, dtype=tf.float16)]\r\n\r\np0 = tf.Variable(params[0])\r\n\r\nout1 = tf.multiply(tf.multiply(p0, inputs), tf.multiply(p0, p0))\r\n\r\nout2 = tf.multiply(p0, tf.multiply(inputs, tf.multiply(p0, p0)))\r\n\r\nfor i in range(min(len(out1), len(out2))):\r\n np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.0001, atol=0.0001, err_msg=f'at checking {i}th')\r\n\r\n\r\nAssertionError: \r\nNot equal to tolerance rtol=0.0001, atol=0.0001\r\nat checking 0th\r\nMismatched elements: 3 / 10 (30%)\r\nMax absolute difference: 0.0004883\r\nMax relative difference: 0.000791\r\n x: array([-0.1956 , -0.6167 , -0.00817, -0.4104 , -0.0771 , -0.07294,\r\n -0.4749 , -0.3445 , -0.501 , -0.1842 ], dtype=float16)\r\n y: array([-0.1956 , -0.617 , -0.00817, -0.4106 , -0.07715, -0.07294,\r\n -0.475 , -0.3445 , -0.501 , -0.1842 ], dtype=float16)\r\n```\r\n\r\nGist: https://gist.github.com/sachinprasadhs/f2cbc7d40e91feb6f50aed1e6ca4e29f", "@sachinprasadhs Hi, Thank you for your response.\r\nI understand that this behavior is caused by a floating-point accuracy problem, which is expected. I have been wondering about the acceptable tolerance level for this floating-point accuracy, given that we have uncovered numerous inconsistent behaviors in TensorFlow. We have only selected and reported cases with high differences in values. I would like to know what the acceptable tolerance in TensorFlow is, so that we can internally filter out some false-positive cases.", "@cantonios , Could you please comment on the above response. Thank you!", "> @sachinprasadhs Hi, Thank you for your response. I understand that this behavior is caused by a floating-point accuracy problem, which is expected. I have been wondering about the acceptable tolerance level for this floating-point accuracy, given that we have uncovered numerous inconsistent behaviors in TensorFlow. We have only selected and reported cases with high differences in values. I would like to know what the acceptable tolerance in TensorFlow is, so that we can internally filter out some false-positive cases.\r\n\r\nYou're using float16 here, which has very limited precision, so the issue is exasperated.\r\n\r\nThe \"acceptable tolerance level\" entirely depends on the sequence of operations you are performing, the dtype, and the magnitudes of your values. You can estimate it yourself: each floating point value is only accurate to within one unit-of-least-precision (ULP), so you can model each value as `x + delta * x` (error is relative to x here due to the way floating-point values are stored - though this is only an approximation, since the error is really relative to `floor(log2(x))`, but that's harder to reason about). The more floating-point operations you do, the more these floating-point errors multiply and accumulate. You can then estimate the expected error for your given set of operations, and you will likely verify that TF (and XLA, and Eigen, pytorch, numpy, etc..) will fall within that.\r\n\r\nThe only thing TF (and XLA) _does_ sometimes do, is for some operations like convolutions and matrix multiplications, we recognize that types like fp16 and bf16 are prone to this kind of bad error accumulation, particularly when multiplying large matrices, so internally we convert them to higher precision and accumulate in higher precision to mitigate the effect.\r\n\r\nThe reported error you are seeing seem reasonable for fp16.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62475\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62475\">No</a>\n" ]	2023-11-24T05:55:05	2023-12-06T18:56:14	2023-12-06T18:56:10	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.16.0-dev20231122 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04.3 ### Mobile device _No response_ ### Python version 3.10.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I've come across a weird issue in TensorFlow. It seems like rearranging the multiplication operations is somehow giving different results. Check out these two models: ```python tf.multiply(tf.multiply(self.p0, inp), tf.multiply(self.p0, self.p0)) ``` ```python tf.multiply(self.p0, tf.multiply(inp, tf.multiply(self.p0, self.p0))) ``` Mathematically, these two tensor operations should be identical since all I did was switch the order of multiplication. But when I run them in TensorFlow with certain inputs, I'm getting different outcomes, which is really strange ### Standalone code to reproduce the issue ```shell import tensorflow as tf import numpy as np params = [ tf.random.uniform(shape=[10, 10], dtype=tf.float16), ] class Model1(tf.keras.Model): def __init__(self): super().__init__() # Tensor objects (with comments for shapes) self.p0 = tf.Variable(params[0]) # [40, 46] float16 # Layers or other Keras model objects @tf.function(jit_compile=True) def __call__(self, inp): # Forward pass logic using TensorFlow operations # inp: [] : float16 out = tf.multiply(tf.multiply(self.p0, inp), tf.multiply(self.p0, self.p0)) # p0 * inp * p0 * p0 return out class Model2(tf.keras.Model): def __init__(self): super().__init__() # Tensor objects (with comments for shapes) self.p0 = tf.Variable(params[0]) # [40, 46] float16 # Layers or other Keras model objects @tf.function(jit_compile=True) def __call__(self, inp): # Forward pass logic using TensorFlow operations # inp: [] : float16 out = tf.multiply(self.p0, tf.multiply(inp, tf.multiply(self.p0, self.p0))) # p0 * inp * (p0 * p0) return out model1 = Model1() model2 = Model2() inputs = [tf.constant(-0.7305, dtype=tf.float16)] print(f'=========RUNNING WITH {inputs}===========') print(inputs[0]) with tf.device('cpu'): tf.config.run_functions_eagerly(True) out1 = model1(inputs) out2 = model2(inputs) print(f'=========eager_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.0001, atol=0.0001, err_msg=f'at checking {i}th') print("XLA_eager does not trigger assertion") except AssertionError as e: print("XLA_eager triggers assertion") print(e) tf.config.run_functions_eagerly(False) out1 = model1(inputs) out2 = model2(inputs) print(f'=========compiled_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.0001, atol=0.0001, err_msg=f'at checking {i}th') print("XLA_complie does not trigger assertion") except AssertionError as e: print("XLA_complie triggers assertion") print(e) ``` ### Relevant log output ```shell =========RUNNING WITH [<tf.Tensor: shape=(), dtype=float16, numpy=-0.7305>]=========== tf.Tensor(-0.7305, shape=(), dtype=float16) =========eager_output(version:2.16.0-dev20231122)================ XLA_eager triggers assertion Not equal to tolerance rtol=0.0001, atol=0.0001 at checking 1th Mismatched elements: 2 / 10 (20%) Max absolute difference: 0.0001221 Max relative difference: 0.0008817 x: array([-2.794e-02, -1.537e-01, -2.107e-02, -1.764e-02, -7.645e-03, -2.986e-01, -1.848e-06, -5.126e-04, -1.383e-01, -5.185e-02], dtype=float16) y: array([-2.794e-02, -1.536e-01, -2.107e-02, -1.764e-02, -7.645e-03, -2.986e-01, -1.848e-06, -5.126e-04, -1.384e-01, -5.185e-02], dtype=float16) 2023-11-24 13:49:22.632282: I external/local_xla/xla/service/service.cc:144] XLA service 0x9170eb0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: 2023-11-24 13:49:22.632327: I external/local_xla/xla/service/service.cc:152] StreamExecutor device (0): NVIDIA GeForce RTX 2070, Compute Capability 7.5 2023-11-24 13:49:22.632338: I external/local_xla/xla/service/service.cc:152] StreamExecutor device (1): NVIDIA GeForce RTX 2070, Compute Capability 7.5 2023-11-24 13:49:22.632347: I external/local_xla/xla/service/service.cc:152] StreamExecutor device (2): NVIDIA GeForce RTX 2070, Compute Capability 7.5 2023-11-24 13:49:22.632357: I external/local_xla/xla/service/service.cc:152] StreamExecutor device (3): NVIDIA GeForce RTX 2070, Compute Capability 7.5 2023-11-24 13:49:22.700845: I external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:456] Loaded cuDNN version 8904 WARNING: All log messages before absl::InitializeLog() is called are written to STDERR I0000 00:00:1700804962.827337 2332797 device_compiler.h:187] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process. =========compiled_output(version:2.16.0-dev20231122)================ XLA_complie triggers assertion Not equal to tolerance rtol=0.0001, atol=0.0001 at checking 1th Mismatched elements: 2 / 10 (20%) Max absolute difference: 0.0001221 Max relative difference: 0.0008817 x: array([-2.794e-02, -1.537e-01, -2.107e-02, -1.764e-02, -7.645e-03, -2.986e-01, -1.848e-06, -5.126e-04, -1.383e-01, -5.185e-02], dtype=float16) y: array([-2.794e-02, -1.536e-01, -2.107e-02, -1.764e-02, -7.645e-03, -2.986e-01, -1.848e-06, -5.126e-04, -1.384e-01, -5.185e-02], dtype=float16) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62475/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62475/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62474	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62474/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62474/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62474/events	https://github.com/tensorflow/tensorflow/issues/62474	2,009,144,533	I_kwDOArmXAs53wRzV	62,474	clEnqueueWriteBuffer To clEnqueueMapBuffer	{ "login": "B-JackMao", "id": 57884144, "node_id": "MDQ6VXNlcjU3ODg0MTQ0", "avatar_url": "https://avatars.githubusercontent.com/u/57884144?v=4", "gravatar_id": "", "url": "https://api.github.com/users/B-JackMao", "html_url": "https://github.com/B-JackMao", "followers_url": "https://api.github.com/users/B-JackMao/followers", "following_url": "https://api.github.com/users/B-JackMao/following{/other_user}", "gists_url": "https://api.github.com/users/B-JackMao/gists{/gist_id}", "starred_url": "https://api.github.com/users/B-JackMao/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/B-JackMao/subscriptions", "organizations_url": "https://api.github.com/users/B-JackMao/orgs", "repos_url": "https://api.github.com/users/B-JackMao/repos", "events_url": "https://api.github.com/users/B-JackMao/events{/privacy}", "received_events_url": "https://api.github.com/users/B-JackMao/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 2671339633, "node_id": "MDU6TGFiZWwyNjcxMzM5NjMz", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TFLiteGpuDelegate", "name": "TFLiteGpuDelegate", "color": "F71F04", "default": false, "description": "TFLite Gpu delegate issue" } ]	closed	false	{ "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false }	[ { "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @B-JackMao, I am having trouble understanding what part of the above is your custom code vs. our code. Is it all your custom code? If any of it is ours, please direct us to where it is so that we may better understand the issue. Reason I ask is we also have EnqueueWriteBuffer: https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/delegates/gpu/cl/cl_command_queue.cc#L139 and it looks different than yours.\r\n\r\nAre you saying the first half of your custom code works, but the 2nd half doesn't?\r\n\r\ni.e. this works?\r\n```\r\nabsl::Status Convert(const TensorObject& input_obj,\r\n const TensorObject& output_obj) override {\r\n \r\n \r\n auto cpu_input = absl::get_if<CpuMemory>(&input_obj);\r\n auto cpu_output = absl::get_if<CpuMemory>(&output_obj);\r\n \r\n if (cpu_input) {\r\n auto texture_output = absl::get_if<OpenClTexture>(&output_obj);\r\n if (texture_output) {\r\n return queue_->EnqueueWriteImage(\r\n texture_output->memobj, int3(region_[0], region_[1], region_[2]),\r\n cpu_input->data, async_);\r\n }\r\n std::cout<<\"write\"<<std::endl;\r\n auto buffer_output = absl::get_if<OpenClBuffer>(&output_obj);\r\n if (buffer_output) {\r\n return queue_->EnqueueWriteBuffer(buffer_output->memobj,\r\n cpu_input->size_bytes,\r\n cpu_input->data, async_);\r\n }\r\n } else if (cpu_output) {\r\n std::cout<<\"read\"<<std::endl;\r\n auto texture_input = absl::get_if<OpenClTexture>(&input_obj);\r\n if (texture_input) {\r\n return queue_->EnqueueReadImage(\r\n texture_input->memobj, int3(region_[0], region_[1], region_[2]),\r\n cpu_output->data, async_);\r\n }\r\n auto buffer_input = absl::get_if<OpenClBuffer>(&input_obj);\r\n if (buffer_input) {\r\n return queue_->EnqueueReadBuffer(buffer_input->memobj,\r\n cpu_output->size_bytes,\r\n cpu_output->data, async_);\r\n }\r\n }\r\n return absl::InternalError(\"Unexpected object\");\r\n }\r\n \r\nabsl::Status CLCommandQueue::EnqueueWriteBuffer(cl_mem memory,\r\n size_t size_in_bytes,\r\n const void* data, bool async) {\r\n const cl_bool blocking = async ? CL_FALSE : CL_TRUE;\r\n std::cout<<\"data write:\"<<(double)data<<std::endl;\r\n auto error_code = clEnqueueWriteBuffer(\r\n queue_, memory, CL_TRUE, 0, size_in_bytes, data, 0, nullptr, nullptr);\r\n if (error_code != CL_SUCCESS) {\r\n return absl::UnknownError(\r\n absl::StrCat(\"Failed to upload data to GPU (clEnqueueWriteBuffer) - \",\r\n CLErrorCodeToString(error_code)));\r\n }\r\n return absl::OkStatus();\r\n}\r\n\r\nabsl::Status CLCommandQueue::EnqueueReadBuffer(cl_mem memory,\r\n size_t size_in_bytes, void* data,\r\n bool async) {\r\n //const cl_bool blocking = async ? CL_FALSE : CL_TRUE;\r\n std::cout<<\"data1 read:\"<<(float)data<<std::endl;\r\n //print_memory(data, size_in_bytes);\r\n auto error_code = clEnqueueReadBuffer(\r\n queue_, memory, CL_TRUE, 0, size_in_bytes, data, 0, nullptr, nullptr);\r\n std::cout<<\"data read:\"<<(char)data<<std::endl;\r\n std::cout<<\"size_in_bytes:\"<<size_in_bytes<<std::endl;\r\n //print_memory(data, size_in_bytes);\r\n if (error_code != CL_SUCCESS) {\r\n return absl::UnknownError(\r\n absl::StrCat(\"Failed to read data from GPU (clEnqueueReadBuffer) - \",\r\n CLErrorCodeToString(error_code)));\r\n }\r\n return absl::OkStatus();\r\n}\r\n```\r\n\r\nbut this doesn't?:\r\n```\r\nabsl::Status Convert(const TensorObject& input_obj,\r\n const TensorObject& output_obj) override {\r\n \r\n auto cpu_input = absl::get_if<CpuMemory>(&input_obj);\r\n auto cpu_output = absl::get_if<CpuMemory>(&output_obj);\r\n if (cpu_input) {\r\n auto buffer_output = absl::get_if<OpenClBuffer>(&output_obj);\r\n if (buffer_output) {\r\n std::cout<<\"map1\"<<std::endl;\r\n return queue_->EnqueueMapBuffer(buffer_output->memobj,cpu_input->size_bytes, cpu_input->data);\r\n }\r\n } else if (cpu_output) {\r\n auto buffer_input = absl::get_if<OpenClBuffer>(&input_obj);\r\n if (buffer_input) {\r\n std::cout<<\"unmap1\"<<std::endl;\r\n return queue_->EnqueueUnMapBuffer(buffer_input->memobj,cpu_output->size_bytes, cpu_output->data);\r\n }\r\n }\r\n \r\n return absl::InternalError(\"Unexpected object\");\r\n }\r\nabsl::Status CLCommandQueue::EnqueueMapBuffer(cl_mem memory,\r\n size_t size_in_bytes,\r\n const void* data) {\r\n std::cout<<\"map\"<<std::endl;\r\n cl_event map_event; \r\n std::cout<<\"data in map:\"<<(double)data<<std::endl;\r\n void* mapped_ptr = clEnqueueMapBuffer(queue_, memory, CL_TRUE, CL_MAP_WRITE, 0, size_in_bytes, 0, nullptr, &map_event, nullptr);// 拷贝数据从映射的指针 \r\n if (mapped_ptr == nullptr) {\r\n return absl::InternalError(\"Failed to map output buffer\");\r\n }\r\n memcpy(mapped_ptr, data, size_in_bytes); // 使用clEnqueueUnmapMemObject来释放映射的指针 \r\n cl_event unmap_event; \r\n cl_int err=clEnqueueUnmapMemObject(queue_, memory, mapped_ptr, 0, nullptr, &unmap_event); // 等待事件完成 \r\n if (err != CL_SUCCESS) {\r\n return absl::InternalError(\"Failed to unmap output buffer\");\r\n }\r\n err =clWaitForEvents(1, &unmap_event); // 返回成功状态 \r\n if (err != CL_SUCCESS) {\r\n return absl::InternalError(\"Failed to wait for unmap event\");\r\n }\r\n // 释放映射和解映射的事件\r\n clReleaseEvent(map_event);\r\n clReleaseEvent(unmap_event);\r\n return absl::OkStatus(); \r\n}\r\nabsl::Status CLCommandQueue::EnqueueUnMapBuffer(cl_mem memory,\r\n size_t size_in_bytes,\r\n void* data) {\r\n std::cout<<\"unmap\"<<std::endl;\r\n cl_event map_event; \r\n std::cout<<\"data1 out map:\"<<(float)data<<std::endl;\r\n void* mapped_ptr = clEnqueueMapBuffer(queue_, memory, CL_TRUE, CL_MAP_READ, 0, size_in_bytes, 0, nullptr, &map_event, nullptr);// 拷贝数据从映射的指针 \r\n if (mapped_ptr == nullptr) {\r\n return absl::InternalError(\"Failed to map input buffer\");\r\n }\r\n \r\n memcpy( data,mapped_ptr,size_in_bytes); // 使用clEnqueueUnmapMemObject来释放映射的指针 \r\n \r\n cl_event unmap_event; \r\n cl_int err =clEnqueueUnmapMemObject(queue_, memory, mapped_ptr, 0, nullptr, &unmap_event); // 等待事件完成 \r\n if (err != CL_SUCCESS) {\r\n return absl::InternalError(\"Failed to unmap input buffer\");\r\n }\r\n err =clWaitForEvents(1, &unmap_event); // 返回成功状态\r\n std::cout<<\"data out map:\"<<(float)data<<std::endl;\r\n std::cout<<\"size_in_bytes:\"<<size_in_bytes<<std::endl;\r\n if (err != CL_SUCCESS) {\r\n return absl::InternalError(\"Failed to wait for unmap event\");\r\n }\r\n // 释放映射和解映射的事件\r\n clReleaseEvent(map_event);\r\n clReleaseEvent(unmap_event); \r\n return absl::OkStatus(); \r\n}\r\n```\r\n\r\nDo you have any example script/data that calls your code that we can use to reproduce your issue? Thanks for anything you can share.", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further." ]	2023-11-24T05:53:06	2023-12-15T01:49:16	2023-12-15T01:49:15	NONE	null	null	null	In TFLite’s GPU delegate, I tried to replace clEnqueueWriteBuffer and clEnqueueReadBuffer with clEnqueueMapBuffer and clEnqueueUnmapMemObject, which means changing the data copying to the data mapping method to pass data, but using clEnqueueWriteBuffer and clEnqueueReadBuffer can correctly recognize the image, but using clEnqueueMapBuffer and clEnqueueUnmapMemObject cannot correctly recognize the image, why is that, the following is the code I modified, what is wrong? `absl::Status Convert(const TensorObject& input_obj, const TensorObject& output_obj) override { auto cpu_input = absl::get_if<CpuMemory>(&input_obj); auto cpu_output = absl::get_if<CpuMemory>(&output_obj); if (cpu_input) { auto texture_output = absl::get_if<OpenClTexture>(&output_obj); if (texture_output) { return queue_->EnqueueWriteImage( texture_output->memobj, int3(region_[0], region_[1], region_[2]), cpu_input->data, async_); } std::cout<<"write"<<std::endl; auto buffer_output = absl::get_if<OpenClBuffer>(&output_obj); if (buffer_output) { return queue_->EnqueueWriteBuffer(buffer_output->memobj, cpu_input->size_bytes, cpu_input->data, async_); } } else if (cpu_output) { std::cout<<"read"<<std::endl; auto texture_input = absl::get_if<OpenClTexture>(&input_obj); if (texture_input) { return queue_->EnqueueReadImage( texture_input->memobj, int3(region_[0], region_[1], region_[2]), cpu_output->data, async_); } auto buffer_input = absl::get_if<OpenClBuffer>(&input_obj); if (buffer_input) { return queue_->EnqueueReadBuffer(buffer_input->memobj, cpu_output->size_bytes, cpu_output->data, async_); } } return absl::InternalError("Unexpected object"); } absl::Status CLCommandQueue::EnqueueWriteBuffer(cl_mem memory, size_t size_in_bytes, const void* data, bool async) { const cl_bool blocking = async ? CL_FALSE : CL_TRUE; std::cout<<"data write:"<<(double)data<<std::endl; auto error_code = clEnqueueWriteBuffer( queue_, memory, CL_TRUE, 0, size_in_bytes, data, 0, nullptr, nullptr); if (error_code != CL_SUCCESS) { return absl::UnknownError( absl::StrCat("Failed to upload data to GPU (clEnqueueWriteBuffer) - ", CLErrorCodeToString(error_code))); } return absl::OkStatus(); } absl::Status CLCommandQueue::EnqueueReadBuffer(cl_mem memory, size_t size_in_bytes, void* data, bool async) { //const cl_bool blocking = async ? CL_FALSE : CL_TRUE; std::cout<<"data1 read:"<<(float)data<<std::endl; //print_memory(data, size_in_bytes); auto error_code = clEnqueueReadBuffer( queue_, memory, CL_TRUE, 0, size_in_bytes, data, 0, nullptr, nullptr); std::cout<<"data read:"<<(char)data<<std::endl; std::cout<<"size_in_bytes:"<<size_in_bytes<<std::endl; //print_memory(data, size_in_bytes); if (error_code != CL_SUCCESS) { return absl::UnknownError( absl::StrCat("Failed to read data from GPU (clEnqueueReadBuffer) - ", CLErrorCodeToString(error_code))); } return absl::OkStatus(); }` clEnqueueReadBuffer can correctly recognize images. `absl::Status Convert(const TensorObject& input_obj, const TensorObject& output_obj) override { auto cpu_input = absl::get_if<CpuMemory>(&input_obj); auto cpu_output = absl::get_if<CpuMemory>(&output_obj); if (cpu_input) { auto buffer_output = absl::get_if<OpenClBuffer>(&output_obj); if (buffer_output) { std::cout<<"map1"<<std::endl; return queue_->EnqueueMapBuffer(buffer_output->memobj,cpu_input->size_bytes, cpu_input->data); } } else if (cpu_output) { auto buffer_input = absl::get_if<OpenClBuffer>(&input_obj); if (buffer_input) { std::cout<<"unmap1"<<std::endl; return queue_->EnqueueUnMapBuffer(buffer_input->memobj,cpu_output->size_bytes, cpu_output->data); } } return absl::InternalError("Unexpected object"); } absl::Status CLCommandQueue::EnqueueMapBuffer(cl_mem memory, size_t size_in_bytes, const void* data) { std::cout<<"map"<<std::endl; cl_event map_event; std::cout<<"data in map:"<<(double)data<<std::endl; void* mapped_ptr = clEnqueueMapBuffer(queue_, memory, CL_TRUE, CL_MAP_WRITE, 0, size_in_bytes, 0, nullptr, &map_event, nullptr);// 拷贝数据从映射的指针 if (mapped_ptr == nullptr) { return absl::InternalError("Failed to map output buffer"); } memcpy(mapped_ptr, data, size_in_bytes); // 使用clEnqueueUnmapMemObject来释放映射的指针 cl_event unmap_event; cl_int err=clEnqueueUnmapMemObject(queue_, memory, mapped_ptr, 0, nullptr, &unmap_event); // 等待事件完成 if (err != CL_SUCCESS) { return absl::InternalError("Failed to unmap output buffer"); } err =clWaitForEvents(1, &unmap_event); // 返回成功状态 if (err != CL_SUCCESS) { return absl::InternalError("Failed to wait for unmap event"); } // 释放映射和解映射的事件 clReleaseEvent(map_event); clReleaseEvent(unmap_event); return absl::OkStatus(); } absl::Status CLCommandQueue::EnqueueUnMapBuffer(cl_mem memory, size_t size_in_bytes, void* data) { std::cout<<"unmap"<<std::endl; cl_event map_event; std::cout<<"data1 out map:"<<(float)data<<std::endl; void* mapped_ptr = clEnqueueMapBuffer(queue_, memory, CL_TRUE, CL_MAP_READ, 0, size_in_bytes, 0, nullptr, &map_event, nullptr);// 拷贝数据从映射的指针 if (mapped_ptr == nullptr) { return absl::InternalError("Failed to map input buffer"); } memcpy( data,mapped_ptr,size_in_bytes); // 使用clEnqueueUnmapMemObject来释放映射的指针 cl_event unmap_event; cl_int err =clEnqueueUnmapMemObject(queue_, memory, mapped_ptr, 0, nullptr, &unmap_event); // 等待事件完成 if (err != CL_SUCCESS) { return absl::InternalError("Failed to unmap input buffer"); } err =clWaitForEvents(1, &unmap_event); // 返回成功状态 std::cout<<"data out map:"<<(float)data<<std::endl; std::cout<<"size_in_bytes:"<<size_in_bytes<<std::endl; if (err != CL_SUCCESS) { return absl::InternalError("Failed to wait for unmap event"); } // 释放映射和解映射的事件 clReleaseEvent(map_event); clReleaseEvent(unmap_event); return absl::OkStatus(); }` clEnqueueMapBuffer can not correctly recognize images. I hope to use clEnqueueMapBuffer for data mapping through TFLite's GPU delegate and be able to correctly identify it. https://stackoverflow.com/questions/77537715/clenqueuewritebuffer-to-clenqueuemapbuffer	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62474/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62474/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62473	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62473/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62473/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62473/events	https://github.com/tensorflow/tensorflow/issues/62473	2,009,140,696	I_kwDOArmXAs53wQ3Y	62,473	clEnqueueWriteBuffer To clEnqueueMapBuffer	{ "login": "B-JackMao", "id": 57884144, "node_id": "MDQ6VXNlcjU3ODg0MTQ0", "avatar_url": "https://avatars.githubusercontent.com/u/57884144?v=4", "gravatar_id": "", "url": "https://api.github.com/users/B-JackMao", "html_url": "https://github.com/B-JackMao", "followers_url": "https://api.github.com/users/B-JackMao/followers", "following_url": "https://api.github.com/users/B-JackMao/following{/other_user}", "gists_url": "https://api.github.com/users/B-JackMao/gists{/gist_id}", "starred_url": "https://api.github.com/users/B-JackMao/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/B-JackMao/subscriptions", "organizations_url": "https://api.github.com/users/B-JackMao/orgs", "repos_url": "https://api.github.com/users/B-JackMao/repos", "events_url": "https://api.github.com/users/B-JackMao/events{/privacy}", "received_events_url": "https://api.github.com/users/B-JackMao/received_events", "type": "User", "site_admin": false }	[ { "id": 1093464312, "node_id": "MDU6TGFiZWwxMDkzNDY0MzEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:others", "name": "type:others", "color": "159b2e", "default": false, "description": "issues not falling in bug, perfromance, support, build and install or feature" } ]	closed	false	{ "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false }	[ { "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false } ]	null	[ "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62473\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62473\">No</a>\n" ]	2023-11-24T05:48:46	2023-11-24T05:53:36	2023-11-24T05:53:33	NONE	null	null	null	### Issue type Others ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version tf 2.9 ### Custom code Yes ### OS platform and distribution Ubuntu18.04 ### Mobile device pixel 4 Android ### Python version 3.8 ### Bazel version 5.0.0 ### GCC/compiler version null ### CUDA/cuDNN version null ### GPU model and memory null ### Current behavior? In TFLite’s GPU delegate, I tried to replace clEnqueueWriteBuffer and clEnqueueReadBuffer with clEnqueueMapBuffer and clEnqueueUnmapMemObject, which means changing the data copying to the data mapping method to pass data, but using clEnqueueWriteBuffer and clEnqueueReadBuffer can correctly recognize the image, but using clEnqueueMapBuffer and clEnqueueUnmapMemObject cannot correctly recognize the image, why is that, the following is the code I modified, what is wrong? `absl::Status Convert(const TensorObject& input_obj, const TensorObject& output_obj) override { auto cpu_input = absl::get_if<CpuMemory>(&input_obj); auto cpu_output = absl::get_if<CpuMemory>(&output_obj); if (cpu_input) { auto texture_output = absl::get_if<OpenClTexture>(&output_obj); if (texture_output) { return queue_->EnqueueWriteImage( texture_output->memobj, int3(region_[0], region_[1], region_[2]), cpu_input->data, async_); } std::cout<<"write"<<std::endl; auto buffer_output = absl::get_if<OpenClBuffer>(&output_obj); if (buffer_output) { return queue_->EnqueueWriteBuffer(buffer_output->memobj, cpu_input->size_bytes, cpu_input->data, async_); } } else if (cpu_output) { std::cout<<"read"<<std::endl; auto texture_input = absl::get_if<OpenClTexture>(&input_obj); if (texture_input) { return queue_->EnqueueReadImage( texture_input->memobj, int3(region_[0], region_[1], region_[2]), cpu_output->data, async_); } auto buffer_input = absl::get_if<OpenClBuffer>(&input_obj); if (buffer_input) { return queue_->EnqueueReadBuffer(buffer_input->memobj, cpu_output->size_bytes, cpu_output->data, async_); } } return absl::InternalError("Unexpected object"); } absl::Status CLCommandQueue::EnqueueWriteBuffer(cl_mem memory, size_t size_in_bytes, const void* data, bool async) { const cl_bool blocking = async ? CL_FALSE : CL_TRUE; std::cout<<"data write:"<<(double)data<<std::endl; auto error_code = clEnqueueWriteBuffer( queue_, memory, CL_TRUE, 0, size_in_bytes, data, 0, nullptr, nullptr); if (error_code != CL_SUCCESS) { return absl::UnknownError( absl::StrCat("Failed to upload data to GPU (clEnqueueWriteBuffer) - ", CLErrorCodeToString(error_code))); } return absl::OkStatus(); } absl::Status CLCommandQueue::EnqueueReadBuffer(cl_mem memory, size_t size_in_bytes, void* data, bool async) { //const cl_bool blocking = async ? CL_FALSE : CL_TRUE; std::cout<<"data1 read:"<<(float)data<<std::endl; //print_memory(data, size_in_bytes); auto error_code = clEnqueueReadBuffer( queue_, memory, CL_TRUE, 0, size_in_bytes, data, 0, nullptr, nullptr); std::cout<<"data read:"<<(char)data<<std::endl; std::cout<<"size_in_bytes:"<<size_in_bytes<<std::endl; //print_memory(data, size_in_bytes); if (error_code != CL_SUCCESS) { return absl::UnknownError( absl::StrCat("Failed to read data from GPU (clEnqueueReadBuffer) - ", CLErrorCodeToString(error_code))); } return absl::OkStatus(); }` `absl::Status Convert(const TensorObject& input_obj, const TensorObject& output_obj) override { auto cpu_input = absl::get_if<CpuMemory>(&input_obj); auto cpu_output = absl::get_if<CpuMemory>(&output_obj); if (cpu_input) { auto buffer_output = absl::get_if<OpenClBuffer>(&output_obj); if (buffer_output) { std::cout<<"map1"<<std::endl; return queue_->EnqueueMapBuffer(buffer_output->memobj,cpu_input->size_bytes, cpu_input->data); } } else if (cpu_output) { auto buffer_input = absl::get_if<OpenClBuffer>(&input_obj); if (buffer_input) { std::cout<<"unmap1"<<std::endl; return queue_->EnqueueUnMapBuffer(buffer_input->memobj,cpu_output->size_bytes, cpu_output->data); } } return absl::InternalError("Unexpected object"); } absl::Status CLCommandQueue::EnqueueMapBuffer(cl_mem memory, size_t size_in_bytes, const void* data) { std::cout<<"map"<<std::endl; cl_event map_event; std::cout<<"data in map:"<<(double)data<<std::endl; void* mapped_ptr = clEnqueueMapBuffer(queue_, memory, CL_TRUE, CL_MAP_WRITE, 0, size_in_bytes, 0, nullptr, &map_event, nullptr);// 拷贝数据从映射的指针 if (mapped_ptr == nullptr) { return absl::InternalError("Failed to map output buffer"); } memcpy(mapped_ptr, data, size_in_bytes); // 使用clEnqueueUnmapMemObject来释放映射的指针 cl_event unmap_event; cl_int err=clEnqueueUnmapMemObject(queue_, memory, mapped_ptr, 0, nullptr, &unmap_event); // 等待事件完成 if (err != CL_SUCCESS) { return absl::InternalError("Failed to unmap output buffer"); } err =clWaitForEvents(1, &unmap_event); // 返回成功状态 if (err != CL_SUCCESS) { return absl::InternalError("Failed to wait for unmap event"); } // 释放映射和解映射的事件 clReleaseEvent(map_event); clReleaseEvent(unmap_event); return absl::OkStatus(); } absl::Status CLCommandQueue::EnqueueUnMapBuffer(cl_mem memory, size_t size_in_bytes, void* data) { std::cout<<"unmap"<<std::endl; cl_event map_event; std::cout<<"data1 out map:"<<(float)data<<std::endl; void* mapped_ptr = clEnqueueMapBuffer(queue_, memory, CL_TRUE, CL_MAP_READ, 0, size_in_bytes, 0, nullptr, &map_event, nullptr);// 拷贝数据从映射的指针 if (mapped_ptr == nullptr) { return absl::InternalError("Failed to map input buffer"); } memcpy( data,mapped_ptr,size_in_bytes); // 使用clEnqueueUnmapMemObject来释放映射的指针 cl_event unmap_event; cl_int err =clEnqueueUnmapMemObject(queue_, memory, mapped_ptr, 0, nullptr, &unmap_event); // 等待事件完成 if (err != CL_SUCCESS) { return absl::InternalError("Failed to unmap input buffer"); } err =clWaitForEvents(1, &unmap_event); // 返回成功状态 std::cout<<"data out map:"<<(float)data<<std::endl; std::cout<<"size_in_bytes:"<<size_in_bytes<<std::endl; if (err != CL_SUCCESS) { return absl::InternalError("Failed to wait for unmap event"); } // 释放映射和解映射的事件 clReleaseEvent(map_event); clReleaseEvent(unmap_event); return absl::OkStatus(); }` I hope to use clEnqueueMapBuffer for data mapping through TFLite's GPU delegate and be able to correctly identify it. ### Standalone code to reproduce the issue ```shell https://stackoverflow.com/questions/77537715/clenqueuewritebuffer-to-clenqueuemapbuffer ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62473/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62473/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62472	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62472/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62472/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62472/events	https://github.com/tensorflow/tensorflow/pull/62472	2,008,784,852	PR_kwDOArmXAs5gQmKL	62,472	Hash Pin docker images	{ "login": "joycebrum", "id": 22223372, "node_id": "MDQ6VXNlcjIyMjIzMzcy", "avatar_url": "https://avatars.githubusercontent.com/u/22223372?v=4", "gravatar_id": "", "url": "https://api.github.com/users/joycebrum", "html_url": "https://github.com/joycebrum", "followers_url": "https://api.github.com/users/joycebrum/followers", "following_url": "https://api.github.com/users/joycebrum/following{/other_user}", "gists_url": "https://api.github.com/users/joycebrum/gists{/gist_id}", "starred_url": "https://api.github.com/users/joycebrum/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/joycebrum/subscriptions", "organizations_url": "https://api.github.com/users/joycebrum/orgs", "repos_url": "https://api.github.com/users/joycebrum/repos", "events_url": "https://api.github.com/users/joycebrum/events{/privacy}", "received_events_url": "https://api.github.com/users/joycebrum/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 1169364458, "node_id": "MDU6TGFiZWwxMTY5MzY0NDU4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:S", "name": "size:S", "color": "adafea", "default": false, "description": "CL Change Size: Small" } ]	open	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "Pinning is a pain when working on these containers. I'm fine with the change as long as you update this PR to alleviate the trouble for developers:\r\n\r\n- Add a dependabot job to keep the shas updated, and add a sha if there is not one\r\n- Add a comment on each affected FROM line that links to the job definition and explains that a sha will be added automatically post-submission", "Hi, I can add the dependabot config file to automatically update the digests on the dockerfiles (which is in fact a security practice too), although dependabot has no feature yet to automatically add hash when there is not one.\r\n\r\nThe renovatebot, although being a third party tool (not GitHub owned), it has an option to [pin digests](https://docs.renovatebot.com/configuration-options/#pindigests) automatically to the given ecosystem (docker images). It would need to be installed on the repo, but only with read permissions, because we can use the [forking renovate](https://github.com/apps/forking-renovate). \r\n\r\nWould you be interested on the renovatebot instead (I can submit a config file for it) or the dependabot only updating the hashes and versions periodically would be enough to alleviate the trouble for developers?", "An additional feature of renovatebot would be the manual update -- when you select a image and ask for an update without having to wait for the next run of renovatebot job ", "Hi @angerson Any update on this PR? Please. Thank you!", "Hi @angerson Any update on this PR? Please. Thank you!", "I've suggested the configuration file for dependabot to show how it would like to update both the actions and the dockerfiles. Dependabot will automatically update the hashes and comments, but it is not able to automatically add a sha when there is none.\r\n", "Hi @angerson Can you please review this PR ? Thank you!", "Hi @joycebrum Can you please resolve conflicts? Thank you!", "Done!", "Hi @MichaelHudgins Can you please review this PR ? Thank you!", "Hi @MichaelHudgins Can you please review this PR ? Thank you!" ]	2023-11-23T21:03:04	2024-06-05T08:20:58	null	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62472", "html_url": "https://github.com/tensorflow/tensorflow/pull/62472", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62472.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62472.patch", "merged_at": null }	Also related to https://github.com/tensorflow/tensorflow/pull/62471, would you consider hash pin the docker images? The security benefit of doing so is that it mitigates the risk of typosquatting attacks since the images are public. If there is a need for them to be updated regularly, I can also submit a .github/dependabot file to update the docker images regularly (weekly or monthly for example). Besides, AFAIUC, the dockerfiles are used for build and tests, which lead to another benefit of hash pinning: reliability and stability. Let me know your thoughts about i. Thanks!	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62472/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62472/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62471	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62471/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62471/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62471/events	https://github.com/tensorflow/tensorflow/pull/62471	2,008,746,221	PR_kwDOArmXAs5gQd5z	62,471	Hash pin GitHub Actions on stale-issues.yml	{ "login": "joycebrum", "id": 22223372, "node_id": "MDQ6VXNlcjIyMjIzMzcy", "avatar_url": "https://avatars.githubusercontent.com/u/22223372?v=4", "gravatar_id": "", "url": "https://api.github.com/users/joycebrum", "html_url": "https://github.com/joycebrum", "followers_url": "https://api.github.com/users/joycebrum/followers", "following_url": "https://api.github.com/users/joycebrum/following{/other_user}", "gists_url": "https://api.github.com/users/joycebrum/gists{/gist_id}", "starred_url": "https://api.github.com/users/joycebrum/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/joycebrum/subscriptions", "organizations_url": "https://api.github.com/users/joycebrum/orgs", "repos_url": "https://api.github.com/users/joycebrum/repos", "events_url": "https://api.github.com/users/joycebrum/events{/privacy}", "received_events_url": "https://api.github.com/users/joycebrum/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169364259, "node_id": "MDU6TGFiZWwxMTY5MzY0MjU5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:XS", "name": "size:XS", "color": "adafea", "default": false, "description": "CL Change Size: Extra Small" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[]	2023-11-23T20:09:01	2023-11-27T16:09:39	2023-11-27T16:09:38	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62471", "html_url": "https://github.com/tensorflow/tensorflow/pull/62471", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62471.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62471.patch", "merged_at": "2023-11-27T16:09:38" }	Hi, moving forward on the contributions mentioned (#61847), I've analyzed the Pinned-Dependencies check results and noticed that only the stale-issues.yml was not hash-pinned. Since the stale-issues.yml has write permissions, it is important to keep it as safe as possible to protect these permissions -- that's how the hash pin would help. Thanks!	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62471/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62471/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62470	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62470/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62470/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62470/events	https://github.com/tensorflow/tensorflow/issues/62470	2,008,700,804	I_kwDOArmXAs53uleE	62,470	Feature request: TensorFlow Hub and HuggingFace integration	{ "login": "AIWithShrey", "id": 24415094, "node_id": "MDQ6VXNlcjI0NDE1MDk0", "avatar_url": "https://avatars.githubusercontent.com/u/24415094?v=4", "gravatar_id": "", "url": "https://api.github.com/users/AIWithShrey", "html_url": "https://github.com/AIWithShrey", "followers_url": "https://api.github.com/users/AIWithShrey/followers", "following_url": "https://api.github.com/users/AIWithShrey/following{/other_user}", "gists_url": "https://api.github.com/users/AIWithShrey/gists{/gist_id}", "starred_url": "https://api.github.com/users/AIWithShrey/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/AIWithShrey/subscriptions", "organizations_url": "https://api.github.com/users/AIWithShrey/orgs", "repos_url": "https://api.github.com/users/AIWithShrey/repos", "events_url": "https://api.github.com/users/AIWithShrey/events{/privacy}", "received_events_url": "https://api.github.com/users/AIWithShrey/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473173272, "node_id": "MDU6TGFiZWw0NzMxNzMyNzI=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:feature", "name": "type:feature", "color": "159b2e", "default": false, "description": "Feature requests" }, { "id": 473184161, "node_id": "MDU6TGFiZWw0NzMxODQxNjE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:support", "name": "type:support", "color": "159b2e", "default": false, "description": "Support issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @AIWithShrey ,\r\n\r\nTfhub already moved to Kaggle models. Please find then announcement of same from [TF-forum.](https://discuss.tensorflow.org/t/tensorflow-hub-is-moving-to-kaggle-models/20770)\r\n\r\nI doubt whether it can be integrated with Hugginface now. Will discuss internally and let you know.\r\n\r\nThanks!", "Hi @AIWithShrey ,\r\n\r\nPlease raise your query at [TF-hub](https://github.com/tensorflow/hub/issues) repo as it should be addressed there. Thanks!", "> Hi @AIWithShrey ,\r\n> \r\n> Please raise your query at [TF-hub](https://github.com/tensorflow/hub/issues) repo as it should be addressed there. Thanks!\r\n\r\nSure! Thanks.", "Hi @AIWithShrey ,\r\n\r\nCould you please close the issue and track at Hub repo?", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62470\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62470\">No</a>\n" ]	2023-11-23T19:08:59	2024-02-13T01:47:20	2024-02-13T01:47:12	NONE	null	null	null	Integrate HuggingFace Hub and TensorFlow Hub for out-of-the-box usage with models and datasets. Support for TensorFlow version of HuggingFace offerings (Transformers, Diffusers, etc.)	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62470/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62470/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62469	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62469/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62469/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62469/events	https://github.com/tensorflow/tensorflow/issues/62469	2,008,658,311	I_kwDOArmXAs53ubGH	62,469	Latest TFLite converter failed on fused operators	{ "login": "psunn", "id": 55463253, "node_id": "MDQ6VXNlcjU1NDYzMjUz", "avatar_url": "https://avatars.githubusercontent.com/u/55463253?v=4", "gravatar_id": "", "url": "https://api.github.com/users/psunn", "html_url": "https://github.com/psunn", "followers_url": "https://api.github.com/users/psunn/followers", "following_url": "https://api.github.com/users/psunn/following{/other_user}", "gists_url": "https://api.github.com/users/psunn/gists{/gist_id}", "starred_url": "https://api.github.com/users/psunn/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/psunn/subscriptions", "organizations_url": "https://api.github.com/users/psunn/orgs", "repos_url": "https://api.github.com/users/psunn/repos", "events_url": "https://api.github.com/users/psunn/events{/privacy}", "received_events_url": "https://api.github.com/users/psunn/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 1661751498, "node_id": "MDU6TGFiZWwxNjYxNzUxNDk4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TFLiteConverter", "name": "TFLiteConverter", "color": "bfdadc", "default": false, "description": "For issues related to TFLite converter" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	closed	false	{ "login": "LakshmiKalaKadali", "id": 149650845, "node_id": "U_kgDOCOt9nQ", "avatar_url": "https://avatars.githubusercontent.com/u/149650845?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LakshmiKalaKadali", "html_url": "https://github.com/LakshmiKalaKadali", "followers_url": "https://api.github.com/users/LakshmiKalaKadali/followers", "following_url": "https://api.github.com/users/LakshmiKalaKadali/following{/other_user}", "gists_url": "https://api.github.com/users/LakshmiKalaKadali/gists{/gist_id}", "starred_url": "https://api.github.com/users/LakshmiKalaKadali/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LakshmiKalaKadali/subscriptions", "organizations_url": "https://api.github.com/users/LakshmiKalaKadali/orgs", "repos_url": "https://api.github.com/users/LakshmiKalaKadali/repos", "events_url": "https://api.github.com/users/LakshmiKalaKadali/events{/privacy}", "received_events_url": "https://api.github.com/users/LakshmiKalaKadali/received_events", "type": "User", "site_admin": false }	[ { "login": "LakshmiKalaKadali", "id": 149650845, "node_id": "U_kgDOCOt9nQ", "avatar_url": "https://avatars.githubusercontent.com/u/149650845?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LakshmiKalaKadali", "html_url": "https://github.com/LakshmiKalaKadali", "followers_url": "https://api.github.com/users/LakshmiKalaKadali/followers", "following_url": "https://api.github.com/users/LakshmiKalaKadali/following{/other_user}", "gists_url": "https://api.github.com/users/LakshmiKalaKadali/gists{/gist_id}", "starred_url": "https://api.github.com/users/LakshmiKalaKadali/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LakshmiKalaKadali/subscriptions", "organizations_url": "https://api.github.com/users/LakshmiKalaKadali/orgs", "repos_url": "https://api.github.com/users/LakshmiKalaKadali/repos", "events_url": "https://api.github.com/users/LakshmiKalaKadali/events{/privacy}", "received_events_url": "https://api.github.com/users/LakshmiKalaKadali/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @psunn ,\r\nI have reproduced the code on tensorflow : 2.15.0. on colab with both CPU and GPU. It is working fine. Please find the [gist](https://colab.research.google.com/gist/LakshmiKalaKadali/b54aa438352215f8921dae753811de17/-62469gpu.ipynb). \r\n\r\n\r\nThank You", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62469\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62469\">No</a>\n" ]	2023-11-23T18:17:52	2023-12-15T01:49:20	2023-12-15T01:49:17	CONTRIBUTOR	null	null	null	The latest TFLite converter seems unable to handle [fused operators ](https://www.tensorflow.org/lite/models/convert/operation_fusion#wrap_the_composite_operation_in_a_tffunction) tflite.converter would be crashed on: <details><summary> code to reproduce </summary> ``` import tensorflow as tf print(f"TensorFlow version: {tf.version.VERSION}") def get_implements_signature(): implements_signature = [ 'name: "exp_sin"', 'attr {key: "tfl_fusable_op" value { b: true } }', ] return " ".join(implements_signature) @tf.function(experimental_implements=get_implements_signature()) def exp_sin(x): x = tf.math.exp(x) x = tf.math.sin(x) return x class custom_example(tf.keras.layers.Layer): def call(self, inputs): return exp_sin(inputs) x_in = x = tf.keras.layers.Input(shape=(10, 10, 3), batch_size=1) x = custom_example()(x) x = tf.keras.layers.Conv2D(filters=8, kernel_size=4)(x) model = tf.keras.Model(inputs=x_in, outputs=x, name="example") model.summary() converter = tf.lite.TFLiteConverter.from_keras_model(model) converter.optimizations = [tf.lite.Optimize.DEFAULT] with open("model_fp32.tflite", "wb") as f: f.write(converter.convert()) ``` </details> <details><summary> error log </summary> ``` 2023-11-23 18:04:30.184288: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CP U instructions in performance-critical operations. To enable the following instructions: AVX2 AVX512F FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. TensorFlow version: 2.16.0-dev20231123 ... 2023-11-23 18:04:32.068288: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:378] Ignored output_format. 2023-11-23 18:04:32.068331: W tensorflow/compiler/mlir/lite/python/tf_tfl_flatbuffer_helpers.cc:381] Ignored drop_control_dependency. 2023-11-23 18:04:32.069026: I tensorflow/cc/saved_model/reader.cc:83] Reading SavedModel from: /tmp/tmp72ka81_f 2023-11-23 18:04:32.069483: I tensorflow/cc/saved_model/reader.cc:51] Reading meta graph with tags { serve } 2023-11-23 18:04:32.069501: I tensorflow/cc/saved_model/reader.cc:146] Reading SavedModel debug info (if present) from: /tmp/tmp72ka81_f 2023-11-23 18:04:32.074256: I tensorflow/compiler/mlir/mlir_graph_optimization_pass.cc:388] MLIR V1 optimization pass is not enabled 2023-11-23 18:04:32.074898: I tensorflow/cc/saved_model/loader.cc:233] Restoring SavedModel bundle. 2023-11-23 18:04:32.114217: I tensorflow/cc/saved_model/loader.cc:217] Running initialization op on SavedModel bundle at path: /tmp/tmp72ka81_f 2023-11-23 18:04:32.120217: I tensorflow/cc/saved_model/loader.cc:316] SavedModel load for tags { serve }; Status: success: OK. Took 51193 microseconds. 2023-11-23 18:04:32.133808: I tensorflow/compiler/mlir/tensorflow/utils/dump_mlir_util.cc:269] disabling MLIR crash reproducer, set env var `MLIR_CRASH_REPRODUCER_DIRECTORY` to enable. loc(fused["ReadVariableOp:", callsite("example_1/conv2d_1/ReadVariableOp@__inference_serving_default_47"("/media/WORK/discovery/NGP_issues/test_models/simple_float_test.py":33:1) at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/tensorflow/lite/python/lite.py":1139:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/tensorflow/lite/python/lite.py":1093:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/tensorflow/lite/python/lite.py":1601:1 at callsite("/home/p ensun01/anaconda3/envs/py39/lib/python3.9/site-packages/tensorflow/lite/python/lite.py":1579:1 at callsite("/home/pensun01/anaconda3/env s/py39/lib/python3.9/site-packages/tensorflow/lite/python/convert_phase.py":205:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/py thon3.9/site-packages/tensorflow/lite/python/lite.py":1502:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages /keras/src/backend/tensorflow/layer.py":57:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/backe nd/tensorflow/layer.py":119:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/utils/traceback_util s.py":118:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/layers/layer.py":830:1 at callsite("/h ome/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/utils/traceback_utils.py":118:1 at callsite("/home/pensun01/anaco nda3/envs/py39/lib/python3.9/site-packages/keras/src/ops/operation.py":42:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3. 9/site-packages/keras/src/utils/traceback_utils.py":157:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/ke ras/src/models/functional.py":188:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/ops/function.p y":153:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/models/functional.py":572:1 at callsite(" /home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/utils/traceback_utils.py":118:1 at callsite("/home/pensun01/ana conda3/envs/py39/lib/python3.9/site-packages/keras/src/layers/layer.py":830:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python 3.9/site-packages/keras/src/utils/traceback_utils.py":118:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/ keras/src/ops/operation.py":42:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/utils/traceback_u tils.py":157:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/layers/convolutional/base_conv.py": 227:1 at callsite("/home/pensun01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/ops/numpy.py":4533:1 at callsite("/home/pens un01/anaconda3/envs/py39/lib/python3.9/site-packages/keras/src/backend/tensorflow/numpy.py":1183:1 at "/home/pensun01/anaconda3/envs/py3 9/lib/python3.9/site-packages/keras/src/backend/tensorflow/core.py":65:1)))))))))))))))))))))))))]): error: missing attribute 'value' LLVM ERROR: Failed to infer result type(s). Aborted (core dumped) ``` </details> This usecase is working on earlier versions, i.g., TensorFlow version: 2.14.1 But it would be failed on recent versions, i.g., TensorFlow version: 2.15.0 ~ TensorFlow version: 2.16.0-dev20231123	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62469/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62469/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62468	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62468/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62468/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62468/events	https://github.com/tensorflow/tensorflow/issues/62468	2,008,568,278	I_kwDOArmXAs53uFHW	62,468	Reducing tensorflowlite_flex shared library with tflite models	{ "login": "aflaischer", "id": 26463144, "node_id": "MDQ6VXNlcjI2NDYzMTQ0", "avatar_url": "https://avatars.githubusercontent.com/u/26463144?v=4", "gravatar_id": "", "url": "https://api.github.com/users/aflaischer", "html_url": "https://github.com/aflaischer", "followers_url": "https://api.github.com/users/aflaischer/followers", "following_url": "https://api.github.com/users/aflaischer/following{/other_user}", "gists_url": "https://api.github.com/users/aflaischer/gists{/gist_id}", "starred_url": "https://api.github.com/users/aflaischer/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/aflaischer/subscriptions", "organizations_url": "https://api.github.com/users/aflaischer/orgs", "repos_url": "https://api.github.com/users/aflaischer/repos", "events_url": "https://api.github.com/users/aflaischer/events{/privacy}", "received_events_url": "https://api.github.com/users/aflaischer/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 1205615612, "node_id": "MDU6TGFiZWwxMjA1NjE1NjEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:%20ubuntu/linux", "name": "subtype: ubuntu/linux", "color": "b619ea", "default": false, "description": "Ubuntu/Linux Build/Installation Issues" }, { "id": 1222092379, "node_id": "MDU6TGFiZWwxMjIyMDkyMzc5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:bazel", "name": "subtype:bazel", "color": "b619ea", "default": false, "description": "Bazel related Build_Installation issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	open	false	{ "login": "terryheo", "id": 2908505, "node_id": "MDQ6VXNlcjI5MDg1MDU=", "avatar_url": "https://avatars.githubusercontent.com/u/2908505?v=4", "gravatar_id": "", "url": "https://api.github.com/users/terryheo", "html_url": "https://github.com/terryheo", "followers_url": "https://api.github.com/users/terryheo/followers", "following_url": "https://api.github.com/users/terryheo/following{/other_user}", "gists_url": "https://api.github.com/users/terryheo/gists{/gist_id}", "starred_url": "https://api.github.com/users/terryheo/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/terryheo/subscriptions", "organizations_url": "https://api.github.com/users/terryheo/orgs", "repos_url": "https://api.github.com/users/terryheo/repos", "events_url": "https://api.github.com/users/terryheo/events{/privacy}", "received_events_url": "https://api.github.com/users/terryheo/received_events", "type": "User", "site_admin": false }	[ { "login": "terryheo", "id": 2908505, "node_id": "MDQ6VXNlcjI5MDg1MDU=", "avatar_url": "https://avatars.githubusercontent.com/u/2908505?v=4", "gravatar_id": "", "url": "https://api.github.com/users/terryheo", "html_url": "https://github.com/terryheo", "followers_url": "https://api.github.com/users/terryheo/followers", "following_url": "https://api.github.com/users/terryheo/following{/other_user}", "gists_url": "https://api.github.com/users/terryheo/gists{/gist_id}", "starred_url": "https://api.github.com/users/terryheo/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/terryheo/subscriptions", "organizations_url": "https://api.github.com/users/terryheo/orgs", "repos_url": "https://api.github.com/users/terryheo/repos", "events_url": "https://api.github.com/users/terryheo/events{/privacy}", "received_events_url": "https://api.github.com/users/terryheo/received_events", "type": "User", "site_admin": false }, { "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @aflaischer, using a non-existent model I got a similar size:\r\n\r\n```\r\n-r-xr-xr-x 1 xxxxxxxx xxxxxxxxxxx 185M Dec 5 22:24 libtensorflowlite_flex.so\r\n```\r\n\r\nIt's possible if no legitimate models are listed it just builds all the flex ops, is it possible for you to share your reference_model.tflite file?", "Hello @pkgoogle \r\n\r\nSure no problem, here is the reference_model.tflite\r\nI put it in a zip file as the tflite extension is not supported as an attachment type.\r\n\r\n[reference_model.zip](https://github.com/tensorflow/tensorflow/files/13579556/reference_model.zip)\r\n", "Hi @aflaischer, thanks for that ... interestingly if I do it with the reference_model it does reduce the size by 2 MB, but that doesn't quite seem right\r\n\r\n```\r\n-r-xr-xr-x 1 xxxxxxx xxxxxxxxxx 183M Dec 6 19:21 libtensorflowlite_flex.so\r\n```\r\n\r\nI looked into the actual build definition: https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/delegates/flex/build_def.bzl#L114 It is taking the models into account ... \r\n\r\nIt is using this binary to list out the flex ops: https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/delegates/flex/build_def.bzl#L55\r\n\r\nSo i tried it manually with your reference file:\r\n\r\n```\r\n./list_flex_ops_main --graphs=reference_model.tflite\r\n[[\"BroadcastGradientArgs\",\"BCastGradArgsOp<int32>\"],[\"ReluGrad\",\"ReluGradOp<CPUDevice, float>\"],[\"Restore\",\"RestoreOp\"],[\"Save\",\"SaveOp\"]]\r\n```\r\n\r\nNot sure why having 4 flex ops would end up reducing the size of the built .so.\r\n\r\n@terryheo can you please take a look? Thanks.", "hello @pkgoogle @terryheo \r\ndo you have some feedback on this topic?\r\n\r\nThanks" ]	2023-11-23T16:52:12	2024-01-04T12:17:29	null	NONE	null	null	null	### Issue type Build/Install ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf2.15 ### Custom code Yes ### OS platform and distribution Linux Ubuntu 20.04.6 LTS ### Mobile device _No response_ ### Python version Python 3.10.13 ### Bazel version bazel 6.1.0 ### GCC/compiler version Ubuntu clang version 17.0.4 ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Hello, I tried to follow the following documentation to reduce the size of the tensorflowlite_flex shared library: https://www.tensorflow.org/lite/guide/reduce_binary_size#advanced_usages_build_custom_cc_shared_libraries So I have made my custom BUILD file with the list of my models: ``` load( "@org_tensorflow//tensorflow/lite/delegates/flex:build_def.bzl", "tflite_flex_shared_library" ) # Shared lib target for convenience, pulls in the standard set of TensorFlow # ops and kernels. The output library name is platform dependent: # - Linux/Android: `libtensorflowlite_flex.so` # - Mac: `libtensorflowlite_flex.dylib` # - Windows: `libtensorflowlite_flex.dll` tflite_flex_shared_library( name = "tensorflowlite_flex", models = [":reference_model.tflite"] ) ``` with tmp created in the root of Tensorflow source repository; ``` tf-docker /tmp/tensorflow > ll tmp/ total 36 -rw-r--r-- 1 root root 1861 Nov 23 15:00 BUILD -rw-r--r-- 1 root root 28152 Nov 23 15:00 reference_model.tflite ``` the compilation is working but I ended up with this size for the tensorflowlite_flex: ``` tf-docker /tmp/tensorflow > bazel build -c opt --cxxopt='--std=c++17' --copt=-Wno-gnu-offsetof-extensions --config=monolithic --host_crosstool_top=@bazel_tools//tools/cpp:toolchain //tmp:tensorflowlite_flex tf-docker /tmp/tensorflow > ll -h bazel-bin/tmp/libtensorflowlite_flex.so -r-xr-xr-x 1 root root 184M Nov 23 16:05 bazel-bin/tmp/libtensorflowlite_flex.so ``` While without using a custom BUILD file: ``` tf-docker /tmp/tensorflow > bazel build -c opt --cxxopt='--std=c++17' --copt=-Wno-gnu-offsetof-extensions --config=monolithic tensorflow/lite/delegates/flex:tensorflowlite_flex tf-docker /tmp/tensorflow > ll -h bazel-bin/tensorflow/lite/delegates/flex/libtensorflowlite_flex.so -r-xr-xr-x 1 root root 184M Nov 23 16:45 bazel-bin/tensorflow/lite/delegates/flex/libtensorflowlite_flex.so ``` I ended up with the same size. I even made a test with an none existing model: ` models = [":model_not_existing.tflite"]` and I get no error but the same size as well. so it is like the model I pass is not taken into account at all. Do you know where the issue could come from? ### Standalone code to reproduce the issue ```shell docker run -it -w /tmp tensorflow/build:2.16-python3.10 git clone --depth 1 --branch r2.15 https://github.com/tensorflow/tensorflow.git cd tensorflow ./configure mkdir -p tmp && touch tmp/BUILD vi tmp/BUILD # add content from above for BUILD file bazel build -c opt --cxxopt='--std=c++17' --copt=-Wno-gnu-offsetof-extensions --config=monolithic --host_crosstool_top=@bazel_tools//tools/cpp:toolchain //tmp:tensorflowlite_flex ll -h bazel-bin/tmp/libtensorflowlite_flex.so ``` ### Relevant log output ```shell -r-xr-xr-x 1 root root 184M Nov 23 16:05 bazel-bin/tmp/libtensorflowlite_flex.so ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62468/reactions", "total_count": 2, "+1": 2, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62468/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62467	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62467/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62467/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62467/events	https://github.com/tensorflow/tensorflow/issues/62467	2,008,557,032	I_kwDOArmXAs53uCXo	62,467	Inconsistency in TensorFlow's Handling of Distributive Properties(`a(b+c) vs ab + ac`) in XLA compiled model	{ "login": "Gwihwan-Go", "id": 83144588, "node_id": "MDQ6VXNlcjgzMTQ0NTg4", "avatar_url": "https://avatars.githubusercontent.com/u/83144588?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Gwihwan-Go", "html_url": "https://github.com/Gwihwan-Go", "followers_url": "https://api.github.com/users/Gwihwan-Go/followers", "following_url": "https://api.github.com/users/Gwihwan-Go/following{/other_user}", "gists_url": "https://api.github.com/users/Gwihwan-Go/gists{/gist_id}", "starred_url": "https://api.github.com/users/Gwihwan-Go/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Gwihwan-Go/subscriptions", "organizations_url": "https://api.github.com/users/Gwihwan-Go/orgs", "repos_url": "https://api.github.com/users/Gwihwan-Go/repos", "events_url": "https://api.github.com/users/Gwihwan-Go/events{/privacy}", "received_events_url": "https://api.github.com/users/Gwihwan-Go/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1133285679, "node_id": "MDU6TGFiZWwxMTMzMjg1Njc5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:xla", "name": "comp:xla", "color": "0052cc", "default": false, "description": "XLA" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false }	[ { "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false } ]	null	[ "@tilakrayal Hey, I was wondering, have you reproduced this issue?", "@Gwihwan-Go,\r\nHi,\r\n\r\nProviding additional outputs does not result in matching outputs. This behavior is expected. Inside the call function you are generating the value for tensor, which in each calls generates different random values because XLA currently ignores TF seeds to random operations which makes the output different for obvious reason. Please refer [known](https://www.tensorflow.org/xla/known_issues#random_number_generation_ignores_tf_seed) issues from XLA section. Also Providing additional outputs does not result in matching outputs. This behavior is expected. Thank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62467\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62467\">No</a>\n" ]	2023-11-23T16:43:09	2024-05-26T01:51:35	2024-05-26T01:51:31	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.14 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04.3 LTS (x86_64) ### Mobile device _No response_ ### Python version 3.10.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I've discovered an inconsistency in TensorFlow's computation, specifically in how distributive properties are handled in tensor operations. Two models, `Model1` and `Model2`, which are mathematically equivalent under the distributive law, are producing different results when compiled with XLA on certain input set(please donwload this [pickle file](https://github.com/GwiHwan-Go/repo/raw/main/0.pickle)). This behavior is also seen on colab with TF 2.14.0 ```python class Model1(tf.keras.Model): @tf.function(jit_compile=True) def __call__(self, inp): v0_0 = tf.abs(inp) v2_0 = tf.negative(inp) v4_0 = tf.add(v0_0, inp) v5_0 = tf.multiply(v2_0, v4_0) return v5_0 # Represents: (abs(inp) + inp) * (-inp) ``` ```python class Model2(tf.keras.Model): @tf.function(jit_compile=True) def __call__(self, inp): v2_0 = tf.negative(inp) v3_0 = tf.abs(inp) v4_0 = tf.multiply(v3_0, v2_0) v5_0 = tf.multiply(inp, v2_0) v6_0 = tf.add(v4_0, v5_0) return v6_0 # Represents: -inp * abs(inp) + inp * -inp ``` ### Standalone code to reproduce the issue ```python ## After download the pickle file from typing import Dict import tensorflow as tf import pickle import os import numpy as np class Model1(tf.keras.Model): def __init__(self): super().__init__() @tf.function(jit_compile=True) def __call__(self, inp): # Forward pass logic using TensorFlow operations # inp: [17, 64, 59, 1, 1] : float32 v0_0 = tf.abs(inp) v2_0 = tf.negative(inp) v4_0 = tf.add(v0_0, inp) v5_0 = tf.multiply(v2_0, v4_0) return v5_0 # (abs(inp) + inp) * (-inp ) class Model2(tf.keras.Model): def __init__(self): super().__init__() @tf.function(jit_compile=True) def __call__(self, inp): # Forward pass logic using TensorFlow operations # inp: [17, 64, 59, 1, 1] : float32 v2_0 = tf.negative(inp) v3_0 = tf.abs(inp) v4_0 = tf.multiply(v3_0, v2_0) v5_0 = tf.multiply(inp, v2_0) v6_0 = tf.add(v4_0, v5_0) return v6_0 # -inp * abs(inp) + inp * -inp model1 = Model1() model2 = Model2() pickle_file_path = YOUR_PICKLE_FILE_PATH if not os.path.exists(pickle_file_path) : print(f'Pickle file not exist') else : with open(pickle_file_path, 'rb') as f : np_arrs1 = pickle.load(f) inputs = [tf.convert_to_tensor(arr) for arr in np_arrs1.values()] with tf.device('cpu'): tf.config.run_functions_eagerly(True) out1 = model1(inputs) out2 = model2(inputs) print(f'=========eager_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.01, err_msg=f'at checking {i}th') print("XLA_eager does not trigger assertion") except AssertionError as e: print("XLA_eager triggers assertion") print(e) tf.config.run_functions_eagerly(False) out1 = model1(inputs) out2 = model2(inputs) print(f'=========compiled_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.01, err_msg=f'at checking {i}th') print("XLA_complie does not trigger assertion") except AssertionError as e: print("XLA_complie triggers assertion") print(e) ``` ### Relevant log output ```shell =========eager_output(version:2.14.0)================ XLA_eager does not trigger assertion =========compiled_output(version:2.14.0)================ XLA_complie triggers assertion Not equal to tolerance rtol=0.01, atol=0 at checking 0th Mismatched elements: 1891 / 3776 (50.1%) Max absolute difference: 4.5960763e-07 Max relative difference: 1. x: array([[[[ 0.000000e+00]], [[ 0.000000e+00]],... y: array([[[[-2.805131e-09]], [[-3.204883e-08]],... ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62467/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62467/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62466	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62466/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62466/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62466/events	https://github.com/tensorflow/tensorflow/issues/62466	2,008,449,236	I_kwDOArmXAs53toDU	62,466	20,000 sleeping threads	{ "login": "ekuznetsov139", "id": 12205429, "node_id": "MDQ6VXNlcjEyMjA1NDI5", "avatar_url": "https://avatars.githubusercontent.com/u/12205429?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ekuznetsov139", "html_url": "https://github.com/ekuznetsov139", "followers_url": "https://api.github.com/users/ekuznetsov139/followers", "following_url": "https://api.github.com/users/ekuznetsov139/following{/other_user}", "gists_url": "https://api.github.com/users/ekuznetsov139/gists{/gist_id}", "starred_url": "https://api.github.com/users/ekuznetsov139/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ekuznetsov139/subscriptions", "organizations_url": "https://api.github.com/users/ekuznetsov139/orgs", "repos_url": "https://api.github.com/users/ekuznetsov139/repos", "events_url": "https://api.github.com/users/ekuznetsov139/events{/privacy}", "received_events_url": "https://api.github.com/users/ekuznetsov139/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 1133285679, "node_id": "MDU6TGFiZWwxMTMzMjg1Njc5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:xla", "name": "comp:xla", "color": "0052cc", "default": false, "description": "XLA" }, { "id": 1463677878, "node_id": "MDU6TGFiZWwxNDYzNjc3ODc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:performance", "name": "type:performance", "color": "159b2e", "default": false, "description": "Performance Issue" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	open	false	{ "login": "ezhulenev", "id": 1174378, "node_id": "MDQ6VXNlcjExNzQzNzg=", "avatar_url": "https://avatars.githubusercontent.com/u/1174378?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ezhulenev", "html_url": "https://github.com/ezhulenev", "followers_url": "https://api.github.com/users/ezhulenev/followers", "following_url": "https://api.github.com/users/ezhulenev/following{/other_user}", "gists_url": "https://api.github.com/users/ezhulenev/gists{/gist_id}", "starred_url": "https://api.github.com/users/ezhulenev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ezhulenev/subscriptions", "organizations_url": "https://api.github.com/users/ezhulenev/orgs", "repos_url": "https://api.github.com/users/ezhulenev/repos", "events_url": "https://api.github.com/users/ezhulenev/events{/privacy}", "received_events_url": "https://api.github.com/users/ezhulenev/received_events", "type": "User", "site_admin": false }	[ { "login": "ezhulenev", "id": 1174378, "node_id": "MDQ6VXNlcjExNzQzNzg=", "avatar_url": "https://avatars.githubusercontent.com/u/1174378?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ezhulenev", "html_url": "https://github.com/ezhulenev", "followers_url": "https://api.github.com/users/ezhulenev/followers", "following_url": "https://api.github.com/users/ezhulenev/following{/other_user}", "gists_url": "https://api.github.com/users/ezhulenev/gists{/gist_id}", "starred_url": "https://api.github.com/users/ezhulenev/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ezhulenev/subscriptions", "organizations_url": "https://api.github.com/users/ezhulenev/orgs", "repos_url": "https://api.github.com/users/ezhulenev/repos", "events_url": "https://api.github.com/users/ezhulenev/events{/privacy}", "received_events_url": "https://api.github.com/users/ezhulenev/received_events", "type": "User", "site_admin": false }, { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[]	2023-11-23T15:29:14	2023-11-30T00:00:37	null	CONTRIBUTOR	null	null	null	Attempting to train a model in a multi-GPU scenario using a recent snapshot of master branch, and discovered that TF creates a huge number of threads, all of which sit idle and do nothing. The exact sequence of operations is as follows: * A collective XLA op is executed. * It calls ResolveDeviceAssignment() https://github.com/tensorflow/tensorflow/blob/master/tensorflow/compiler/jit/kernels/xla_ops.cc#L616 * Which calls BaseCollectiveExecutor::CompleteParamsAsync() with a timeout of 1000 seconds https://github.com/tensorflow/tensorflow/blob/master/tensorflow/compiler/tf2xla/xla_helpers.cc#L147 * Which calls SchedNonBlockingClosureAfter() https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/common_runtime/base_collective_executor.cc#L361 * Which creates a thread which sleeps for the specified period of time (1000 s) and then executes the supplied closure function https://github.com/tensorflow/tensorflow/blob/master/third_party/xla/third_party/tsl/tsl/platform/default/env.cc#L181 The problem is essentially that SchedNonBlockingClosureAfter() is implemented in an exceedingly wasteful way, because its author assumed it to be something used once in a blue moon, and now it's being called for every collective XLA op (i.e. multiple calls for every step during training). I suspect that this massive pool of idle threads is what causes some of my training runs to crash after a while, with "terminate called after throwing an instance of 'std::system_error' / terminate called recursively / what(): Resource temporarily unavailable".	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62466/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62466/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62465	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62465/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62465/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62465/events	https://github.com/tensorflow/tensorflow/issues/62465	2,008,065,205	I_kwDOArmXAs53sKS1	62,465	tf.linalg.logdet outputs NaN on positive definite matrixs	{ "login": "drewshark", "id": 128925028, "node_id": "U_kgDOB689ZA", "avatar_url": "https://avatars.githubusercontent.com/u/128925028?v=4", "gravatar_id": "", "url": "https://api.github.com/users/drewshark", "html_url": "https://github.com/drewshark", "followers_url": "https://api.github.com/users/drewshark/followers", "following_url": "https://api.github.com/users/drewshark/following{/other_user}", "gists_url": "https://api.github.com/users/drewshark/gists{/gist_id}", "starred_url": "https://api.github.com/users/drewshark/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/drewshark/subscriptions", "organizations_url": "https://api.github.com/users/drewshark/orgs", "repos_url": "https://api.github.com/users/drewshark/repos", "events_url": "https://api.github.com/users/drewshark/events{/privacy}", "received_events_url": "https://api.github.com/users/drewshark/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" } ]	closed	false	{ "login": "cantonios", "id": 2538739, "node_id": "MDQ6VXNlcjI1Mzg3Mzk=", "avatar_url": "https://avatars.githubusercontent.com/u/2538739?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cantonios", "html_url": "https://github.com/cantonios", "followers_url": "https://api.github.com/users/cantonios/followers", "following_url": "https://api.github.com/users/cantonios/following{/other_user}", "gists_url": "https://api.github.com/users/cantonios/gists{/gist_id}", "starred_url": "https://api.github.com/users/cantonios/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cantonios/subscriptions", "organizations_url": "https://api.github.com/users/cantonios/orgs", "repos_url": "https://api.github.com/users/cantonios/repos", "events_url": "https://api.github.com/users/cantonios/events{/privacy}", "received_events_url": "https://api.github.com/users/cantonios/received_events", "type": "User", "site_admin": false }	[ { "login": "cantonios", "id": 2538739, "node_id": "MDQ6VXNlcjI1Mzg3Mzk=", "avatar_url": "https://avatars.githubusercontent.com/u/2538739?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cantonios", "html_url": "https://github.com/cantonios", "followers_url": "https://api.github.com/users/cantonios/followers", "following_url": "https://api.github.com/users/cantonios/following{/other_user}", "gists_url": "https://api.github.com/users/cantonios/gists{/gist_id}", "starred_url": "https://api.github.com/users/cantonios/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cantonios/subscriptions", "organizations_url": "https://api.github.com/users/cantonios/orgs", "repos_url": "https://api.github.com/users/cantonios/repos", "events_url": "https://api.github.com/users/cantonios/events{/privacy}", "received_events_url": "https://api.github.com/users/cantonios/received_events", "type": "User", "site_admin": false }, { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@drewshark Though I was able to replicate the issue reported [here](https://colab.research.google.com/gist/sushreebarsa/231e65b2b8a44f66dfe9a821d02d2dc9/62465.ipynb), could you try to use the tf.math.log(tf.linalg.det(x)) formula instead of tf.linalg.logdet. Please let us know if it helps?\r\nThank you!", "Hi @sushreebarsa Thanks for your reply. tf.math.log(tf.linalg.det(x)) works correctly. However, it would be nice of you to point the possible root causes of tf.linalg.logdet, I am willing to offer help in fixing this issue.", "@drewshark The algorithm used by tf.linalg.logdet might also have limitations in handling certain types of positive definite matrices. For example, certain matrix structures might trigger corner cases within the algorithm, leading to unexpected NaN outputs. `tf.math.log(tf.linalg.det(x)) ` can be used alternatively. Please check the [doc](https://www.tensorflow.org/api_docs/python/tf/linalg/logdet) to know more on the limitations as well. Thank you!", "Hi @sushreebarsa \r\n\r\nI can accept the case where tf.linalg.logdet has some limitations. However, I still cannot understand what's the limitation is by checking the [doc](https://www.tensorflow.org/api_docs/python/tf/linalg/logdet). Particularly, it would be more reliable if I can know when should I use tf.math.log(tf.linalg.det(x)) and when should I use tf.linalg.logdet. Moreover, I notice in the doc, tf.linalg.logdet should be \"Equivalent to numpy.linalg.slogdet, although no sign is returned since only hermitian positive definite matrices are supported.\" However, np.linalg.slogdet can correctly handle this matrix instead of outputting NaN like tf.linalg.logdet. Here is the comparison code for your reference\r\n\r\n```\r\nimport os\r\nimport tensorflow as tf\r\nimport numpy as np\r\nx = [[0.53077246, 0.39622202], [0.71875037, 0.548092]]\r\ntf_matrix = tf.constant(x, dtype='float64')\r\nprint(tf.linalg.logdet(tf_matrix)) # output NaN\r\nprint(np.linalg.slogdet(tf_matrix)) # output (1.0, -5.094982205337635)\r\n```", "@drewshark , \r\n\r\nThanks for reporting the issue.\r\n\r\nAfter spending some time on the issue to debug, below are my analysis.\r\n\r\n`NaN` value from `tf.linalg.logdet` is basically from `Cholesky decomposition`, since it is generating NaN value, the subsequent calculations are passed NaN values and ultimately resulting in `NaN` output. \r\n\r\nFor `Cholesky` decomposition, the input matrix has to be positive definite, which is handled in the `Numpy` `Cholesky` implementation. \r\n\r\nTo break it down:\r\n\r\nThe formula for tf.linalg.logdet is \r\n*log det(A) = 2sum(log(real(diag(C))))** source [link](https://github.com/tensorflow/tensorflow/blob/4dacf3f368eb7965e9b5c3bbdd5193986081c3b2/tensorflow/python/ops/linalg/linalg_impl.py#L68-L99).\r\nwhere C is the cholesky decomposition of A.\r\n\r\nAs per the above formula if we calculate step by step, it results in the below output, as I explained above.\r\n\r\n```\r\nimport numpy as np\r\nimport tensorflow as tf\r\n\r\nC = np.array([[0.53077246, 0.39622202],\r\n [0.71875037, 0.548092]])\r\n\r\ncholesky = tf.linalg.cholesky(C)\r\nprint(\"cholesky: \" ,cholesky)\r\n_cholesky: tf.Tensor(\r\n[[ nan 4.71671851e-310]\r\n [ nan nan]], shape=(2, 2), dtype=float64)_\r\n\r\n\r\ndiag_part = tf.linalg.diag_part(cholesky)\r\nprint(\"diag_part: \", diag_part)\r\n_diag_part: tf.Tensor([nan nan], shape=(2,), dtype=float64)_\r\n\r\nlog = tf.math.log(diag_part)\r\nprint(\"log: \",log)\r\n_log: tf.Tensor([nan nan], shape=(2,), dtype=float64)_\r\n\r\n\r\n\r\nreduce_sum = tf.math.reduce_sum(log)\r\nprint(\"reduce_sum: \", reduce_sum)\r\n_reduce_sum: tf.Tensor(nan, shape=(), dtype=float64)_\r\n\r\nans = 2 * reduce_sum\r\nprint(ans)\r\n _tf.Tensor(nan, shape=(), dtype=float64)_\r\n```\r\n\r\nHere is the Gist with the above example https://gist.github.com/sachinprasadhs/2c7ea55525aee7a6ad80971c9524020f\r\n\r\n@cantonios, do you think we need to handle in `cholesky` to check if it is a positive definite matrix as `Numpy` is handling.\r\nAlso, I could not figure out why `tf.linalg.cholesky(C)` in the linked Gist is generating random numbers for each run.\r\n", "According to the [docs](https://www.tensorflow.org/api_docs/python/tf/linalg/logdet), `tf.linalg.logdet`:\r\n\r\n> Computes log of the determinant of a hermitian positive definite matrix.\r\n\r\nYour matrix is not hermitian (i.e. symmetric).", "Hi @sachinprasadhs , I appreciate your analysis, I understand how tf.linalg.logdet is computed now. However, I find tf.linalg.logdet still outputs incorrect result even if tf.linalg.cholesky works correctly. Here is my input that triggers the issue:\r\nC = [[0.73077246, 0.39622202], [0.21875037, 0.548092]]\r\n\r\nIn short, given this matrix, tf.linalg.logdet's output is largely different from np.linalg.slogdet and tf.math.log(tf.linalg.det(C)). tf.linalg.logdet's output is -1.0421976 while the other two output -1.1588184. Note that tf.linalg.cholesky's computation result is correct.\r\n\r\nHere is my analysis code:\r\n\r\n```\r\nimport numpy as np\r\nimport tensorflow as tf\r\n\r\nC = [[0.73077246, 0.39622202], [0.21875037, 0.548092]]\r\n\r\ncholesky = tf.linalg.cholesky(C)\r\nprint(\"cholesky: \" ,cholesky)\r\n# cholesky: tf.Tensor(\r\n# [[0.85485226 0. ]\r\n# [0.2558926 0.6947021 ]], shape=(2, 2), dtype=float32)\r\n\r\n\r\ndiag_part = tf.linalg.diag_part(cholesky)\r\nprint(\"diag_part: \", diag_part)\r\n# diag_part: tf.Tensor([0.85485226 0.6947021 ], shape=(2,), dtype=float32)\r\n\r\nlog = tf.math.log(diag_part)\r\nprint(\"log: \",log)\r\n# log: tf.Tensor([-0.15682662 -0.36427218], shape=(2,), dtype=float32)\r\n\r\nreduce_sum = tf.math.reduce_sum(log)\r\nprint(\"reduce_sum: \", reduce_sum)\r\n# reduce_sum: tf.Tensor(-0.5210988, shape=(), dtype=float32)\r\n\r\nans = 2 * reduce_sum\r\nprint(f\"res from step-by-step computation in tf.linalg.logdet\", ans)\r\n# res from step-by-step computation in tf.linalg.logdet tf.Tensor(-1.0421976, shape=(), dtype=float32)\r\nprint(f\"res from np.linalg.slogdet\", np.linalg.slogdet(C)[-1])\r\n# res from np.linalg.slogdet -1.1588183662659097\r\nprint(f\"res from tf.linalg.logdet\", tf.linalg.logdet(C))\r\n# res from tf.linalg.logdet tf.Tensor(-1.0421976, shape=(), dtype=float32)\r\nprint(f\"res from log(det(x)))\", tf.math.log(tf.linalg.det(C)))\r\n# res from log(det(x))) tf.Tensor(-1.1588184, shape=(), dtype=float32)\r\n```\r\n\r\nHere are my suggestions:\r\n\r\n- maybe the formula of tf.linalg.logdet is unstable to some corner case values, thus leading to such a large discrepancy. I want to check how other library implements this function but I fail to locate their implementations. It would be better if you can offer some references on this aspect.\r\n- For @cantonios's comment,\r\n> Your matrix is not hermitian (i.e. symmetric).\r\n\r\nI am not an expert in this part so I cannot easily identify whether my input is a hermitian matrix or not, therefore, when giving the input to tf.linalg.logdet, I can't easily tell if tf.linalg.logdet's output is reliable or not. If would be better if tf.linalg.logdet can some additional checking to tell me I cannot send such input to tf.linalg.logdet, or I cannot rely on its result for this input.", "@drewshark your matrix is not symmetric [[hermitian matrix wiki]](https://en.m.wikipedia.org/wiki/Hermitian_matrix). Therefore your input matrix, as well as the result of a cholesky decomposition, is garbage. The decomposition is failing, leading to nans in the result, which is how we always handle failed decompositions. We purposely don't fail with a crash for bad inputs because sometimes these errors are sporadic in real life models, and users prefer to explicitly handle nans instead of crashing.\r\n\r\nWe could potentially use a different decomposition in the case of non symmetric matrices. That would be expanding functionality though. As documented, your input is invalid.", "Hi @cantonios , thanks a lot for your explanation, I can understand outputting NaN result is the way you handle failed decomposition. I am fine with the way you handle my original input: x = [[0.53077246, 0.39622202], [0.71875037, 0.548092]], which is not a hermitian matrix.\r\n\r\nAnother problem of mine is that, during analyzing this issue, I find another input: C = [[0.73077246, 0.39622202], [0.21875037, 0.548092]], which is a slight adjustment of original one. Although this input is not hermitian matrix, tf.linalg.logdet neither raise a message nor outputting nan, instead, tf.linalg.logdet just outputs an incorrect result\r\n\r\n```\r\nimport numpy as np\r\nimport tensorflow as tf\r\n\r\nC = [[0.73077246, 0.39622202], [0.21875037, 0.548092]]\r\n\r\nprint(f\"res from np.linalg.slogdet\", np.linalg.slogdet(C)[-1])\r\n# res from np.linalg.slogdet -1.1588183662659097\r\n\r\nprint(f\"res from tf.linalg.logdet\", tf.linalg.logdet(C))\r\n# res from tf.linalg.logdet tf.Tensor(-1.0421976, shape=(), dtype=float32)\r\n\r\nprint(f\"res from log(det(x)))\", tf.math.log(tf.linalg.det(C)))\r\n# res from log(det(x))) tf.Tensor(-1.1588184, shape=(), dtype=float32)\r\n```\r\n\r\nYou can reproduce the issue by running above code, which simply send the input to different implementation of logdet. tf.linalg.logdet(C)'s result is largely different from np.linalg.slogdet(C) and tf.math.log(tf.linalg.det(C)). \r\n\r\nFor some users who are not familiar with the usage of tf.linalg.logdet or who are not careful enough, they may directly trust the result of tf.linalg.logdet on this input or may get confused like I previously did. Therefore, I think an additional check on tf.linalg.logdet's input will definitely help users to better use this function. Additionally, it would be much better if you can expand this functionality to handle non symmetric matrices.\r\n\r\nBy all means, I appreciate your attention in addressing my issue.", "> Another problem of mine is that, during analyzing this issue, I find another input: C = [[0.73077246, 0.39622202], [0.21875037, 0.548092]], which is a slight adjustment of original one. Although this input is not hermitian matrix, tf.linalg.logdet neither raise a message nor outputting nan, instead, tf.linalg.logdet just outputs an incorrect result\r\n\r\nAgain, your matrix fails the precondition of the function you are using. Garbage in = garbage out. This time the decomposition didn't technically fail, as the routine assumed the matrix was symmetric only read values from the lower half. However, the results are still bogus. We don't specifically check inputs for them being symmetric or positive definite, since doing so is usually the same amount of work as doing the actual computation. It's up to you to ensure you're using the function properly.\r\n\r\nNote that scipy works the same. If you give it your matrix, `scipy.linalg.cholesky` will produce a result assuming your matrix was symmetric:\r\n```python\r\n>>> C = np.array([[0.73077246, 0.39622202], [0.21875037, 0.548092]])\r\n>>> U = scipy.linalg.cholesky(C)\r\n>>> print(\"U =\\n \", U)\r\n>>> print(\"U^T * U =\\n\", np.transpose(U) @ U)\r\n\r\nU: \r\n[[0.8548523 0.46349763]\r\n [0. 0.57728844]]\r\nU^T * U = \r\n[[0.73077246 0.39622202]\r\n [0.39622202 0.548092 ]]\r\n```\r\nNote the decomposition only reproduce the upper triangular part of your original input.", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62465\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62465\">No</a>\n" ]	2023-11-23T11:49:50	2023-12-20T01:42:55	2023-12-20T01:42:46	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.16.0-dev20231123 ### Custom code Yes ### OS platform and distribution _No response_ ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Given the matrix `x = [[0.53077246, 0.39622202], [0.71875037, 0.548092]]`, whose determinant is positive (0.006). Its log determinant is NaN returned by tf.linalg.logdet. In contrast, if I use tf.math.log(tf.linalg.det(x)) instead, it can outputs normal value. Additionally, if I change the x a little bit (`x = [[0.73077246, 0.39622202], [0.21875037, 0.548092]]`), there is a large difference between tf.linalg.logdet and tf.math.log(tf.linalg.det()). ### Standalone code to reproduce the issue ```shell import os import tensorflow as tf x = [[0.53077246, 0.39622202], [0.71875037, 0.548092]] tf_matrix = tf.constant(x, dtype='float64') print(tf.linalg.logdet(tf_matrix), tf.linalg.det(tf_matrix), tf.math.log(tf.linalg.det(tf_matrix))) # tf.Tensor(nan, shape=(), dtype=float64) tf.Tensor(0.006127415669172652, shape=(), dtype=float64) tf.Tensor(-5.094982205337635, shape=(), dtype=float64) ``` ``` import os import tensorflow as tf x = [[0.73077246, 0.39622202], [0.21875037, 0.548092]] tf_matrix = tf.constant(x, dtype='float64') print(tf.linalg.logdet(tf_matrix), tf.math.log(tf.linalg.det(tf_matrix))) # tf.Tensor(-1.0421975094041425, shape=(), dtype=float64) tf.Tensor(-1.1588183662659097, shape=(), dtype=float64) ``` ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62465/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62465/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62464	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62464/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62464/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62464/events	https://github.com/tensorflow/tensorflow/issues/62464	2,008,014,785	I_kwDOArmXAs53r9_B	62,464	OneDNN Verbose not visible in tfv2.14.0 and above.	{ "login": "vishwascm", "id": 40938161, "node_id": "MDQ6VXNlcjQwOTM4MTYx", "avatar_url": "https://avatars.githubusercontent.com/u/40938161?v=4", "gravatar_id": "", "url": "https://api.github.com/users/vishwascm", "html_url": "https://github.com/vishwascm", "followers_url": "https://api.github.com/users/vishwascm/followers", "following_url": "https://api.github.com/users/vishwascm/following{/other_user}", "gists_url": "https://api.github.com/users/vishwascm/gists{/gist_id}", "starred_url": "https://api.github.com/users/vishwascm/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/vishwascm/subscriptions", "organizations_url": "https://api.github.com/users/vishwascm/orgs", "repos_url": "https://api.github.com/users/vishwascm/repos", "events_url": "https://api.github.com/users/vishwascm/events{/privacy}", "received_events_url": "https://api.github.com/users/vishwascm/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1104829434, "node_id": "MDU6TGFiZWwxMTA0ODI5NDM0", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:mkl", "name": "comp:mkl", "color": "0052cc", "default": false, "description": "MKL related issues" }, { "id": 1205615612, "node_id": "MDU6TGFiZWwxMjA1NjE1NjEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:%20ubuntu/linux", "name": "subtype: ubuntu/linux", "color": "b619ea", "default": false, "description": "Ubuntu/Linux Build/Installation Issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @vishwascm ,\r\n\r\nIf the Ops are executed by XLA its unlikely to get oneDNN logs. Could you please check this [source](https://github.com/tensorflow/tensorflow/issues/61322#issuecomment-1645122803) for detailed discussion on similar issue.", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62464\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62464\">No</a>\n" ]	2023-11-23T11:16:51	2023-12-04T04:26:39	2023-12-04T04:26:35	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source binary ### TensorFlow version v2.14.0 ### Custom code No ### OS platform and distribution ubuntu 22.04.1 on aarch64 ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? pip install tensorflow==2.14.0 export TF_ENABLE_ONEDNN_OPTS=1 export ONEDNN_VERBOSE=1 Run a example to use onednn will give following output: oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. That's it no OneDNN logs visible. But if I use v2.13.0 pip install tensorflow==2.13.0 export TF_ENABLE_ONEDNN_OPTS=1 export ONEDNN_VERBOSE=1 Run a example to use onednn will give following output: Experimental oneDNN custom operations are on. If you experience issues, please turn them off by setting the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. onednn_verbose,info,oneDNN v2.7.3 (commit N/A) onednn_verbose,info,cpu,runtime:threadpool,nthr:32 onednn_verbose,info,cpu,isa:AArch64 SVE (256 bits) onednn_verbose,info,gpu,runtime:none onednn_verbose,info,prim_template:operation,engine,primitive,implementation,prop_kind,memory_descriptors,attributes,auxiliary,problem_desc,exec_time onednn_verbose,exec,cpu,softmax_v2,ref:any,forward_inference,src_f32::blocked:abcd:f0 dst_f32::blocked:abcd:f0,,alg:softmax_accurate axis:3,1x12x2x2,0.141846 ### Standalone code to reproduce the issue ```shell pip install tensorflow==2.14.0 export TF_ENABLE_ONEDNN_OPTS=1 export ONEDNN_VERBOSE=1 Run any neural network containing softmax layer: ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62464/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62464/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62463	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62463/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62463/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62463/events	https://github.com/tensorflow/tensorflow/issues/62463	2,007,775,615	I_kwDOArmXAs53rDl_	62,463	Inconsistency in TensorFlow's Handling of Associative Operations: `(a+b)c - (a+b)c vs. - (ac + bc) + ac + bc`	{ "login": "Gwihwan-Go", "id": 83144588, "node_id": "MDQ6VXNlcjgzMTQ0NTg4", "avatar_url": "https://avatars.githubusercontent.com/u/83144588?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Gwihwan-Go", "html_url": "https://github.com/Gwihwan-Go", "followers_url": "https://api.github.com/users/Gwihwan-Go/followers", "following_url": "https://api.github.com/users/Gwihwan-Go/following{/other_user}", "gists_url": "https://api.github.com/users/Gwihwan-Go/gists{/gist_id}", "starred_url": "https://api.github.com/users/Gwihwan-Go/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Gwihwan-Go/subscriptions", "organizations_url": "https://api.github.com/users/Gwihwan-Go/orgs", "repos_url": "https://api.github.com/users/Gwihwan-Go/repos", "events_url": "https://api.github.com/users/Gwihwan-Go/events{/privacy}", "received_events_url": "https://api.github.com/users/Gwihwan-Go/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097545817, "node_id": "MDU6TGFiZWwxMDk3NTQ1ODE3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:apis", "name": "comp:apis", "color": "0052cc", "default": false, "description": "Highlevel API related issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@tilakrayal Hey, I was wondering, have you replicated this issue?", "@sachinprasadhs,\r\nI was able to reproduce the issue on tensorflow v2.14, v2.15 and tf-nightly. Kindly find the gist of it [here](https://colab.research.google.com/gist/tilakrayal/4bf126a7a0bec88ef2dbc07deeb60828/untitled.ipynb)." ]	2023-11-23T09:15:07	2023-11-29T23:57:58	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.16.0-dev20231122 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04.3 ### Mobile device _No response_ ### Python version 3.10.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? This bug is also seen on nightly version of tf(`2.16.0-dev20231122`). I have two TensorFlow models (`Model1` and `Model2`) that are semantically the same but produce different outputs when executed. This discrepancy might indicate an underlying issue in TensorFlow's execution or optimization paths, particularly with XLA compilation. ```python class Model1(tf.keras.Model): def __init__(self): super().__init__() # Tensor objects (with comments for shapes) self.p0 = tf.Variable(params[0]) # [1] float32 self.p1 = tf.Variable(params[1]) # [] float32 # Layers or other Keras model objects @tf.function(jit_compile=True) def __call__(self, inp): # Forward pass logic using TensorFlow operations # inp: [52, 33, 4] : float32 v9_0 = tf.add(self.p1, self.p0) v5_0 = tf.multiply(v9_0, inp) v0_0 = tf.negative(v5_0) v11_0 = tf.add(v5_0, v0_0) return v11_0 ``` ```python class Model2(tf.keras.Model): def __init__(self): super().__init__() # Tensor objects (with comments for shapes) self.p0 = tf.Variable(params[0]) # [1] float32 self.p1 = tf.Variable(params[1]) # [] float32 # Layers or other Keras model objects @tf.function(jit_compile=True) def __call__(self, inp): # Forward pass logic using TensorFlow operations # inp: [52, 33, 4] : float32 v10_0 = tf.multiply(self.p0, inp) v12_0 = tf.multiply(self.p1, inp) v13_0 = tf.add(v12_0, v10_0) v17_0 = tf.negative(v13_0) v38_0 = tf.add(v17_0, v10_0) v39_0 = tf.add(v38_0, v12_0) return v39_0 ``` Given that Model1 and Model2 are semantically the same (i.e.,` model1 = (p0+p1)inp - (p0+p1)inp` and `model2 = - (p0inp + p1inp) + p0inp + p1inp`), I would expect them to produce identical outputs for the same input. However, the models are yielding different results, which is unexpected and suggests a potential issue in TensorFlow's computation or optimization mechanisms. ### Standalone code to reproduce the issue ```shell from typing import Dict import tensorflow as tf import os import numpy as np params = [ tf.random.uniform(shape=[1], dtype=tf.float32), tf.random.uniform(shape=[], dtype=tf.float32), ] class Model1(tf.keras.Model): def __init__(self): super().__init__() # Tensor objects (with comments for shapes) self.p0 = tf.Variable(params[0]) # [1] float32 self.p1 = tf.Variable(params[1]) # [] float32 # Layers or other Keras model objects @tf.function(jit_compile=True) def __call__(self, inp): # Forward pass logic using TensorFlow operations # inp: [52, 33, 4] : float32 v9_0 = tf.add(self.p1, self.p0) v5_0 = tf.multiply(v9_0, inp) v0_0 = tf.negative(v5_0) v11_0 = tf.add(v5_0, v0_0) return v11_0 class Model2(tf.keras.Model): def __init__(self): super().__init__() # Tensor objects (with comments for shapes) self.p0 = tf.Variable(params[0]) # [1] float32 self.p1 = tf.Variable(params[1]) # [] float32 # Layers or other Keras model objects @tf.function(jit_compile=True) def __call__(self, inp): # Forward pass logic using TensorFlow operations # inp: [52, 33, 4] : float32 v10_0 = tf.multiply(self.p0, inp) v12_0 = tf.multiply(self.p1, inp) v13_0 = tf.add(v12_0, v10_0) v17_0 = tf.negative(v13_0) v38_0 = tf.add(v17_0, v10_0) v39_0 = tf.add(v38_0, v12_0) return v39_0 inputs = [ tf.random.uniform(shape=[10,10,10], dtype=tf.float32), ] model1 = Model1() model2 = Model2() tf.config.run_functions_eagerly(True) out1 = model1(inputs) out2 = model2(inputs) print(f'=========eager_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.01, err_msg=f'at checking {i}th') print("XLA_eager does not trigger assertion") except AssertionError as e: print("XLA_eager triggers assertion") print(e) tf.config.run_functions_eagerly(False) out1 = model1(inputs) out2 = model2(inputs) print(f'=========compiled_output(version:{tf.__version__})================') try : for i in range(min(len(out1),len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.01, err_msg=f'at checking {i}th') print("XLA_complie does not trigger assertion") except AssertionError as e: print("XLA_complie triggers assertion") print(e) ``` ### Relevant log output ```shell 2023-11-23 17:14:24.344078: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. 2023-11-23 17:14:25.779158: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT 2023-11-23 17:14:28.121908: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 5558 MB memory: -> device: 0, name: NVIDIA GeForce RTX 2070, pci bus id: 0000:02:00.0, compute capability: 7.5 2023-11-23 17:14:28.122693: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:1 with 4932 MB memory: -> device: 1, name: NVIDIA GeForce RTX 2070, pci bus id: 0000:04:00.0, compute capability: 7.5 2023-11-23 17:14:28.123338: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:2 with 4080 MB memory: -> device: 2, name: NVIDIA GeForce RTX 2070, pci bus id: 0000:83:00.0, compute capability: 7.5 2023-11-23 17:14:28.123938: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1929] Created device /job:localhost/replica:0/task:0/device:GPU:3 with 6062 MB memory: -> device: 3, name: NVIDIA GeForce RTX 2070, pci bus id: 0000:84:00.0, compute capability: 7.5 =========eager_output(version:2.16.0-dev20231122)================ XLA_eager triggers assertion Not equal to tolerance rtol=0.01, atol=0 at checking 0th Mismatched elements: 66 / 100 (66%) Max absolute difference: 5.9604645e-08 Max relative difference: 1. x: array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],... y: array([[ 5.960464e-08, -2.980232e-08, 0.000000e+00, -1.490116e-08, -2.980232e-08, 5.960464e-08, 0.000000e+00, 5.960464e-08, 7.450581e-09, -2.980232e-08],... 2023-11-23 17:14:28.641380: I external/local_xla/xla/service/service.cc:144] XLA service 0x950adf0 initialized for platform CUDA (this does not guarantee that XLA will be used). Devices: 2023-11-23 17:14:28.641420: I external/local_xla/xla/service/service.cc:152] StreamExecutor device (0): NVIDIA GeForce RTX 2070, Compute Capability 7.5 2023-11-23 17:14:28.641432: I external/local_xla/xla/service/service.cc:152] StreamExecutor device (1): NVIDIA GeForce RTX 2070, Compute Capability 7.5 2023-11-23 17:14:28.641444: I external/local_xla/xla/service/service.cc:152] StreamExecutor device (2): NVIDIA GeForce RTX 2070, Compute Capability 7.5 2023-11-23 17:14:28.641455: I external/local_xla/xla/service/service.cc:152] StreamExecutor device (3): NVIDIA GeForce RTX 2070, Compute Capability 7.5 2023-11-23 17:14:28.679907: I external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:456] Loaded cuDNN version 8904 WARNING: All log messages before absl::InitializeLog() is called are written to STDERR I0000 00:00:1700730868.755509 2277729 device_compiler.h:187] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process. =========compiled_output(version:2.16.0-dev20231122)================ XLA_complie triggers assertion Not equal to tolerance rtol=0.01, atol=0 at checking 0th Mismatched elements: 66 / 100 (66%) Max absolute difference: 5.9604645e-08 Max relative difference: 1. x: array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0.],... y: array([[ 5.960464e-08, -2.980232e-08, 0.000000e+00, -1.490116e-08, -2.980232e-08, 5.960464e-08, 0.000000e+00, 5.960464e-08, 7.450581e-09, -2.980232e-08],... ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62463/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62463/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62462	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62462/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62462/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62462/events	https://github.com/tensorflow/tensorflow/issues/62462	2,007,716,683	I_kwDOArmXAs53q1NL	62,462	TFLIte interpreter crashes on Pixel 6 by not finding OpenCL lib	{ "login": "raydenvoldeskine", "id": 1571937, "node_id": "MDQ6VXNlcjE1NzE5Mzc=", "avatar_url": "https://avatars.githubusercontent.com/u/1571937?v=4", "gravatar_id": "", "url": "https://api.github.com/users/raydenvoldeskine", "html_url": "https://github.com/raydenvoldeskine", "followers_url": "https://api.github.com/users/raydenvoldeskine/followers", "following_url": "https://api.github.com/users/raydenvoldeskine/following{/other_user}", "gists_url": "https://api.github.com/users/raydenvoldeskine/gists{/gist_id}", "starred_url": "https://api.github.com/users/raydenvoldeskine/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/raydenvoldeskine/subscriptions", "organizations_url": "https://api.github.com/users/raydenvoldeskine/orgs", "repos_url": "https://api.github.com/users/raydenvoldeskine/repos", "events_url": "https://api.github.com/users/raydenvoldeskine/events{/privacy}", "received_events_url": "https://api.github.com/users/raydenvoldeskine/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473184161, "node_id": "MDU6TGFiZWw0NzMxODQxNjE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:support", "name": "type:support", "color": "159b2e", "default": false, "description": "Support issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" } ]	closed	false	{ "login": "LakshmiKalaKadali", "id": 149650845, "node_id": "U_kgDOCOt9nQ", "avatar_url": "https://avatars.githubusercontent.com/u/149650845?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LakshmiKalaKadali", "html_url": "https://github.com/LakshmiKalaKadali", "followers_url": "https://api.github.com/users/LakshmiKalaKadali/followers", "following_url": "https://api.github.com/users/LakshmiKalaKadali/following{/other_user}", "gists_url": "https://api.github.com/users/LakshmiKalaKadali/gists{/gist_id}", "starred_url": "https://api.github.com/users/LakshmiKalaKadali/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LakshmiKalaKadali/subscriptions", "organizations_url": "https://api.github.com/users/LakshmiKalaKadali/orgs", "repos_url": "https://api.github.com/users/LakshmiKalaKadali/repos", "events_url": "https://api.github.com/users/LakshmiKalaKadali/events{/privacy}", "received_events_url": "https://api.github.com/users/LakshmiKalaKadali/received_events", "type": "User", "site_admin": false }	[ { "login": "LakshmiKalaKadali", "id": 149650845, "node_id": "U_kgDOCOt9nQ", "avatar_url": "https://avatars.githubusercontent.com/u/149650845?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LakshmiKalaKadali", "html_url": "https://github.com/LakshmiKalaKadali", "followers_url": "https://api.github.com/users/LakshmiKalaKadali/followers", "following_url": "https://api.github.com/users/LakshmiKalaKadali/following{/other_user}", "gists_url": "https://api.github.com/users/LakshmiKalaKadali/gists{/gist_id}", "starred_url": "https://api.github.com/users/LakshmiKalaKadali/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LakshmiKalaKadali/subscriptions", "organizations_url": "https://api.github.com/users/LakshmiKalaKadali/orgs", "repos_url": "https://api.github.com/users/LakshmiKalaKadali/repos", "events_url": "https://api.github.com/users/LakshmiKalaKadali/events{/privacy}", "received_events_url": "https://api.github.com/users/LakshmiKalaKadali/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @raydenvoldeskine ,\r\n\r\nPlease add the following to the Android manifest in order to detect GPU delegate.\r\n\r\n```\r\n<uses-library android:name=\"libOpenCL.so\"\r\n android:required=\"false\"/>\r\n\r\n<uses-library android:name=\"libOpenCL-pixel.so\"\r\n android:required=\"false\"/>\r\n```\r\nPlease let us know if it helps you.\r\n\r\nThank you", "Thank you for a really fast feedback. The trick did work on Pixel 6 device. Could you pls also explain the general situation: \r\n\r\n-> Is that an official approach to use with TFLtie on Android? If yes, why it is not relfected in initial configuration instructions - or did I miss something? \r\n-> what about other devices - Pixels or not Pixels, which would not have openCL (or maybe have an incompatible version)? I want to get more confidence that after we made this fix, the app still works on other devices. \r\n", "Hi\r\n@raydenvoldeskine ,\r\n\r\nResponse1: Yes, it's an official approach . Applications that target Android S+ require explicit declaration of\r\n any referenced vendor-provided libraries. Ref[ doc](https://github.com/tensorflow/tensorflow/blob/7b8dd63247d2d62254c3021c1f1188a60eb5eb85/tensorflow/lite/java/AndroidManifestGpuApi.xml).\r\n\r\nResponse2: This fix works on android 12 and above. OpenCL allows developers to write code that can run on different devices, regardless of the manufacturer or architecture, by providing a common set of commands and a runtime environment that can execute the code on the target device.\r\nCheck these related issues also. [#60720](https://github.com/tensorflow/tensorflow/issues/60720), [#48001](https://github.com/tensorflow/tensorflow/issues/48001), [#50394](https://github.com/tensorflow/tensorflow/issues/50394). Please let us know if u have any new observations\r\n\r\n\r\n\r\nThank You", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62462\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62462\">No</a>\n" ]	2023-11-23T08:41:02	2023-12-12T01:49:45	2023-12-12T01:49:43	NONE	null	null	null	### Issue type Support ### Have you reproduced the bug with TensorFlow Nightly? No ### Source binary ### TensorFlow version 2.5.0 ### Custom code No ### OS platform and distribution Android ### Mobile device Pixel 6, Android 13 ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? TFlite Interpreter by instantiation crashes the app with message when started with GPU (see below) Using NNAPI does not cause the crash, yet the initialisation of interpreter takes about 20 s which is hardly acceptable. Note that running Benchmark tool on same device with same model, on GPU works. ### Standalone code to reproduce the issue ```shell Not possible ``` ### Relevant log output ```shell `java.lang.IllegalArgumentException: Internal error: Failed to run on the given Interpreter: Can not open OpenCL library on this device - dlopen failed: library "libOpenCL.so" not found` ` TfLiteGpuDelegate Invoke: [GL_INVALID_VALUE]: A numeric argument is out of range.: glDispatchCompute in tensorflow/lite/delegates/gpu/gl/gl_program.cc:217` ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62462/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62462/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62461	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62461/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62461/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62461/events	https://github.com/tensorflow/tensorflow/issues/62461	2,007,644,031	I_kwDOArmXAs53qjd_	62,461	Inconsistent Outputs from `tf.linalg.eigh` in XLA-Compiled Model	{ "login": "Gwihwan-Go", "id": 83144588, "node_id": "MDQ6VXNlcjgzMTQ0NTg4", "avatar_url": "https://avatars.githubusercontent.com/u/83144588?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Gwihwan-Go", "html_url": "https://github.com/Gwihwan-Go", "followers_url": "https://api.github.com/users/Gwihwan-Go/followers", "following_url": "https://api.github.com/users/Gwihwan-Go/following{/other_user}", "gists_url": "https://api.github.com/users/Gwihwan-Go/gists{/gist_id}", "starred_url": "https://api.github.com/users/Gwihwan-Go/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Gwihwan-Go/subscriptions", "organizations_url": "https://api.github.com/users/Gwihwan-Go/orgs", "repos_url": "https://api.github.com/users/Gwihwan-Go/repos", "events_url": "https://api.github.com/users/Gwihwan-Go/events{/privacy}", "received_events_url": "https://api.github.com/users/Gwihwan-Go/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1133285679, "node_id": "MDU6TGFiZWwxMTMzMjg1Njc5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:xla", "name": "comp:xla", "color": "0052cc", "default": false, "description": "XLA" }, { "id": 5508003926, "node_id": "LA_kwDOArmXAs8AAAABSE14Vg", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.13", "name": "TF 2.13", "color": "B13ACB", "default": false, "description": "For issues related to Tensorflow 2.13" } ]	closed	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@sachinprasadhs I was able to replicate the issue reported [here](https://colab.research.google.com/gist/sushreebarsa/93603cff7886c43d8b1f17b2d7f37906/62461.ipynb). Please have a look at this issue. \r\nThank you!", "Hi, \r\n\r\nSince the code you have mentioned has random number generation, the outputs can not be guaranteed for each run.\r\nI have tried by setting seed and enabling op determinism and still produces random results.\r\n```\r\ntf.random.set_seed(42)\r\ntf.config.experimental.enable_op_determinism() \r\n```", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "Hi @sachinprasadhs,\r\nIs this issue expected behavior? I believe the random behavior is not directly related to the issue.", "Your matrices are not hermitian (`A.conjugate().transpose() != A`), so the results are garbage, even in the eager case. We don't check this precondition since checking is just as expensive as the operation itself. Behavior is therefore undefined.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62461\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62461\">No</a>\n" ]	2023-11-23T07:46:46	2024-01-09T21:10:24	2024-01-09T21:10:21	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.13.0 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04.3 ### Mobile device _No response_ ### Python version 3.10.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I've encountered an issue where `tf.linalg.eigh`, when used in an XLA-compiled TensorFlow model, produces different outputs for the same input on consecutive runs. ### Standalone code to reproduce the issue ```python import tensorflow as tf import numpy as np class Model1(tf.keras.Model): def __init__(self): super().__init__() @tf.function(jit_compile=True) def __call__(self, v4_0): v0_0, _ = tf.linalg.eigh(v4_0) return v0_0 inputs = [ tf.complex(tf.random.uniform([1, 48, 1, 1], dtype=tf.float32), tf.random.uniform([1, 48, 1, 1], dtype=tf.float32)), ] model1 = Model1() with tf.device('cpu'): # Test in eager execution mode tf.config.run_functions_eagerly(True) out1 = model1(inputs) out2 = model1(inputs) print(f'=========eager_output(version:{tf.__version__})================') try: for i in range(min(len(out1), len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.01, err_msg=f'at checking {i}th') print("XLA_eager does not trigger assertion") except AssertionError as e: print("XLA_eager triggers assertion") print(e) # Test in compiled mode tf.config.run_functions_eagerly(False) out1 = model1(inputs) out2 = model1(inputs) print(f'=========compiled_output(version:{tf.__version__})================') try: for i in range(min(len(out1), len(out2))): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.01, err_msg=f'at checking {i}th') print("XLA_compile does not trigger assertion") except AssertionError as e: print("XLA_compile triggers assertion") print(e) ``` ### Relevant log output ```shell Skipping registering GPU devices... =========eager_output(version:2.13.0)================ XLA_eager does not trigger assertion 2023-11-23 07:38:29.271178: I tensorflow/compiler/xla/service/service.cc:168] XLA service 0x561bb6eca560 initialized for platform Host (this does not guarantee that XLA will be used). Devices: 2023-11-23 07:38:29.271197: I tensorflow/compiler/xla/service/service.cc:176] StreamExecutor device (0): Host, Default Version 2023-11-23 07:38:29.278907: I ./tensorflow/compiler/jit/device_compiler.h:186] Compiled cluster using XLA! This line is logged at most once for the lifetime of the process. =========compiled_output(version:2.13.0)================ XLA_complie triggers assertion Not equal to tolerance rtol=0.01, atol=0 at checking 0th x and y nan location mismatch: x: array([[ 3.354893e-01+1.480364e-01j], [ 4.999712e-01+2.206396e-01j], [ 4.584086e-02+1.555587e-01j],... y: array([[ 3.354893e-01+1.480364e-01j], [ 4.999712e-01+2.206396e-01j], [ 4.584086e-02+1.555587e-01j],... ``` ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62461/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62461/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62460	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62460/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62460/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62460/events	https://github.com/tensorflow/tensorflow/issues/62460	2,007,493,278	I_kwDOArmXAs53p-qe	62,460	keras.Model.fit does not work correctly with generator and sparse categorical crossentropy loss	{ "login": "oonisim", "id": 15814603, "node_id": "MDQ6VXNlcjE1ODE0NjAz", "avatar_url": "https://avatars.githubusercontent.com/u/15814603?v=4", "gravatar_id": "", "url": "https://api.github.com/users/oonisim", "html_url": "https://github.com/oonisim", "followers_url": "https://api.github.com/users/oonisim/followers", "following_url": "https://api.github.com/users/oonisim/following{/other_user}", "gists_url": "https://api.github.com/users/oonisim/gists{/gist_id}", "starred_url": "https://api.github.com/users/oonisim/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/oonisim/subscriptions", "organizations_url": "https://api.github.com/users/oonisim/orgs", "repos_url": "https://api.github.com/users/oonisim/repos", "events_url": "https://api.github.com/users/oonisim/events{/privacy}", "received_events_url": "https://api.github.com/users/oonisim/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "## Workaround\r\n\r\nThe answer by [innat](https://stackoverflow.com/users/9215780/innat) worked as in https://stackoverflow.com/a/77535234/4281353.\r\n\r\n> The behaviour that is found with metrics=[\"accuracy\"] for using sparse target vectors seems like a potential bug in the API. According to the [doc](https://github.com/keras-team/keras/blob/68f9af408a1734704746f7e6fa9cfede0d6879d8/keras/engine/training.py#L684-L689), the string identifier accuracy should be converted to appropriate loss instance.\r\n> In you case, you need to use ```keras.metrics.SparseCategoricalAccuracy(name='accuracy')``` specifically to make it work.\r\n\r\n```\r\nmodel.compile(\r\n optimizer=Adam(learning_rate=learning_rate, beta_1=0.9, beta_2=0.999, epsilon=1e-08),\r\n #metrics=[\"accuracy\"]\r\n loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False),\r\n metrics=[tf.keras.metrics.SparseCategoricalAccuracy(name='accuracy')])\r\nmodel.summary()\r\n```", "@oonisim ,\r\n\r\nThe reason for this behaviour might also be due to numeric instability issue which happens with `logits=False`. For more details on this you can refer SO source here [1](https://stackoverflow.com/questions/57253841/from-logits-true-and-from-logits-false-get-different-training-result-for-tf-loss/71365020#71365020) & [2](https://stackoverflow.com/questions/52125924/why-does-sigmoid-crossentropy-of-keras-tensorflow-have-low-precision/52126567#52126567). If I change `logits=True` in loss function and set activation to `linear` (default) in last layer of model the model trains with gradual improvement and callbacks were not called in this case. Refer attached [gist](https://colab.sandbox.google.com/gist/SuryanarayanaY/475147b479d919eb1902d21b43d93109/62460.ipynb).\r\n\r\nHowever as per your [comment](https://github.com/tensorflow/tensorflow/issues/62460#issuecomment-1824105048) above replacing `metrics=[\"accuracy\"]` with `metrics=[tf.keras.metrics.SparseCategoricalAccuracy(name='accuracy')])` works even better.But as per docstring below of trainer.py from keras3 there seems some issue with the conversion of `accuracy` to `SparseCategoricalAccuracy`.\r\n\r\n```\r\n metrics=[['accuracy'], ['accuracy', 'mse']]`\r\n or `metrics=['accuracy', ['accuracy', 'mse']]`. When you pass\r\n the strings 'accuracy' or 'acc', we convert this to one of\r\n `keras.metrics.BinaryAccuracy`,\r\n `keras.metrics.CategoricalAccuracy`,\r\n `keras.metrics.SparseCategoricalAccuracy` based on the\r\n shapes of the targets and of the model output.\r\n```\r\n\r\nhttps://github.com/keras-team/keras/blob/866b745ebdafb29248f6a0946fbed1ce1cbcba90/keras/trainers/trainer.py#L88", "Hi @oonisim ,\r\n\r\nUpdate... This issue seems fixed in keras-nightly(3.0.0.dev2023112703) already. I have tested with keras3 and its working fine. The `accuracy` metric automatically converted into `SparseCategoricalAccuracy` from [here](https://github.com/keras-team/keras/blob/master/keras/trainers/compile_utils.py#L66C4-L66C4).\r\n\r\nYou can refer the attached [gist](https://colab.sandbox.google.com/gist/SuryanarayanaY/930e1e82bfc512d0f9474a0273188f75/62460_keras-nigthly.ipynb#scrollTo=26NP2vgKWe7s) of same. Could you verify the behaviour with keras-nightly and confirm ?", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62460\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62460\">No</a>\n" ]	2023-11-23T05:05:03	2023-12-12T01:49:50	2023-12-12T01:49:47	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source binary ### TensorFlow version 2.14.1 ### Custom code No ### OS platform and distribution Ubuntu 22.04 LTS ### Mobile device _No response_ ### Python version 3.10.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version 11.8 ### GPU model and memory _No response_ ### Current behavior? ```tf.keras.Model.fit(x=generator)``` does not work correctly with ```SparseCategoricalCrossentropy```/```sparce_categorical_crossentropy``` loss function with a generator as training data. The same symptom reported in [Accuracy killed when using ImageDataGenerator TensorFlow Keras][1]. Please advise if this behaviour is as expected or please point out if code is incorrect. Code excerpt. Entire code at the bottom. ``` # -------------------------------------------------------------------------------- # CIFAR 10 # -------------------------------------------------------------------------------- USE_SPARCE_LABEL = True (x_train, y_train), (x_test, y_test) = tf.keras.datasets.cifar10.load_data() x_train, x_test = x_train / 255.0, x_test / 255.0 x_train, x_validation, y_train, y_validation = train_test_split( x_train, y_train, test_size=0.2, random_state=42 ) # One Hot Encoding the labels when USE_SPARCE_LABEL is False if not USE_SPARCE_LABEL: y_train = keras.utils.to_categorical(y_train, NUM_CLASSES) y_validation = keras.utils.to_categorical(y_validation, NUM_CLASSES) y_test = keras.utils.to_categorical(y_test, NUM_CLASSES) # -------------------------------------------------------------------------------- # Model # -------------------------------------------------------------------------------- model: Model = Model( inputs=inputs, outputs=outputs, name="cifar10" ) # -------------------------------------------------------------------------------- # Compile # -------------------------------------------------------------------------------- if USE_SPARCE_LABEL: loss_fn=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False) # <--- cause incorrect behavior else: loss_fn=tf.keras.losses.CategoricalCrossentropy(from_logits=False) learning_rate = 1e-3 model.compile( optimizer=Adam(learning_rate=learning_rate, beta_1=0.9, beta_2=0.999, epsilon=1e-08), loss=loss_fn, # <---- sparse categorical causes the incorrect behavior metrics=["accuracy"] ) # -------------------------------------------------------------------------------- # Train # -------------------------------------------------------------------------------- batch_size = 16 number_of_epochs = 10 def data_label_generator(x, y): def _f(): index = 0 length = len(x) try: while True: yield x[index:index+batch_size], y[index:index+batch_size] index = (index + batch_size) % length except StopIteration: return return _f earlystop_callback = tf.keras.callbacks.EarlyStopping( patience=5, restore_best_weights=True, monitor='val_accuracy' ) steps_per_epoch = len(y_train) // batch_size validation_steps = (len(y_validation) // batch_size) - 1 # To avoid run out of data for validation history = model.fit( x=data_label_generator(x_train, y_train)(), # <--- Generator batch_size=batch_size, epochs=number_of_epochs, verbose=1, validation_data=data_label_generator(x_validation, y_validation)(), shuffle=True, steps_per_epoch=steps_per_epoch, validation_steps=validation_steps, validation_batch_size=batch_size, callbacks=[ earlystop_callback ] ) ``` Symptom Using Sparse Index as the labels with and SparseCategoricalCrossentropy as the loss function. The accuracy values got unstable and low, causing early stop. ----------- 2500/2500 [...] - 24s 8ms/step - loss: 1.4824 - accuracy: 0.0998 - val_loss: 1.1893 - val_accuracy: 0.1003 Epoch 2/10 2500/2500 [...] - 21s 8ms/step - loss: 1.0730 - accuracy: 0.1010 - val_loss: 0.8896 - val_accuracy: 0.0832 Epoch 3/10 2500/2500 [...] - 20s 8ms/step - loss: 0.9272 - accuracy: 0.1016 - val_loss: 0.9150 - val_accuracy: 0.0720 Epoch 4/10 2500/2500 [...] - 20s 8ms/step - loss: 0.7987 - accuracy: 0.1019 - val_loss: 0.8087 - val_accuracy: 0.0864 Epoch 5/10 2500/2500 [...] - 20s 8ms/step - loss: 0.7081 - accuracy: 0.1012 - val_loss: 0.8707 - val_accuracy: 0.0928 Epoch 6/10 2500/2500 [...] - 21s 8ms/step - loss: 0.6056 - accuracy: 0.1019 - val_loss: 0.7688 - val_accuracy: 0.0851 ----------- Using One Hot Encoding as the labels and ```CategoricalCrossentropy``` as the loss function (```USE_SPARSE_LABEL=True```). Work as expected. ``` 2500/2500 [...] - 24s 8ms/step - loss: 1.4146 - accuracy: 0.4997 - val_loss: 1.0906 - val_accuracy: 0.6105 Epoch 2/10 2500/2500 [...] - 21s 9ms/step - loss: 1.0306 - accuracy: 0.6375 - val_loss: 0.9779 - val_accuracy: 0.6532 Epoch 3/10 2500/2500 [...] - 22s 9ms/step - loss: 0.8780 - accuracy: 0.6925 - val_loss: 0.8194 - val_accuracy: 0.7127 Epoch 4/10 2500/2500 [...] - 21s 8ms/step - loss: 0.7641 - accuracy: 0.7315 - val_loss: 0.9330 - val_accuracy: 0.7014 Epoch 5/10 2500/2500 [...] - 21s 8ms/step - loss: 0.6797 - accuracy: 0.7614 - val_loss: 0.7908 - val_accuracy: 0.7311 Epoch 6/10 2500/2500 [...] - 21s 9ms/step - loss: 0.6182 - accuracy: 0.7841 - val_loss: 0.7371 - val_accuracy: 0.7533 Epoch 7/10 2500/2500 [...] - 21s 9ms/step - loss: 0.4981 - accuracy: 0.8217 - val_loss: 0.8221 - val_accuracy: 0.7373 Epoch 8/10 2500/2500 [...] - 22s 9ms/step - loss: 0.4363 - accuracy: 0.8437 - val_loss: 0.7865 - val_accuracy: 0.7525 Epoch 9/10 2500/2500 [...] - 23s 9ms/step - loss: 0.3962 - accuracy: 0.8596 - val_loss: 0.8198 - val_accuracy: 0.7505 Epoch 10/10 2500/2500 [...] - 22s 9ms/step - loss: 0.3463 - accuracy: 0.8776 - val_loss: 0.8472 - val_accuracy: 0.7512 ``` ## Code ``` import numpy as np import tensorflow as tf from tensorflow import keras from keras import ( __version__ ) from keras.layers import ( Layer, Normalization, Conv2D, MaxPooling2D, BatchNormalization, Dense, Flatten, Dropout, Reshape, Activation, ReLU, LeakyReLU, ) from keras.models import ( Model, ) from keras.layers import ( Layer ) from keras.optimizers import ( Adam ) from sklearn.model_selection import train_test_split print("TensorFlow version: {}".format(tf.__version__)) tf.keras.__version__ = __version__ print("Keras version: {}".format(tf.keras.__version__)) # -------------------------------------------------------------------------------- # CIFAR 10 # -------------------------------------------------------------------------------- NUM_CLASSES = 10 INPUT_SHAPE = (32, 32, 3) USE_SPARCE_LABEL = False # Setting False make it work as expected (x_train, y_train), (x_test, y_test) = tf.keras.datasets.cifar10.load_data() x_train, x_test = x_train / 255.0, x_test / 255.0 x_train, x_validation, y_train, y_validation = train_test_split( x_train, y_train, test_size=0.2, random_state=42 ) # One Hot Encoding the labels if not USE_SPARCE_LABEL: y_train = keras.utils.to_categorical(y_train, NUM_CLASSES) y_validation = keras.utils.to_categorical(y_validation, NUM_CLASSES) y_test = keras.utils.to_categorical(y_test, NUM_CLASSES) # -------------------------------------------------------------------------------- # Model # -------------------------------------------------------------------------------- inputs = tf.keras.Input( name='image', shape=INPUT_SHAPE, dtype=tf.float32 ) x = Conv2D( filters=32, kernel_size=(3, 3), strides=(1, 1), padding="same", activation='relu', input_shape=INPUT_SHAPE )(inputs) x = BatchNormalization()(x) x = Conv2D( filters=64, kernel_size=(3, 3), strides=(1, 1), padding="same", activation='relu' )(x) x = MaxPooling2D( pool_size=(2, 2) )(x) x = Dropout(0.20)(x) x = Conv2D( filters=128, kernel_size=(3, 3), strides=(1, 1), padding="same", activation='relu' )(x) x = BatchNormalization()(x) x = MaxPooling2D( pool_size=(2, 2) )(x) x = Dropout(0.20)(x) x = Flatten()(x) x = Dense(300, activation="relu")(x) x = BatchNormalization()(x) x = Dropout(0.20)(x) x = Dense(200, activation="relu")(x) outputs = Dense(NUM_CLASSES, activation="softmax")(x) model: Model = Model( inputs=inputs, outputs=outputs, name="cifar10" ) # -------------------------------------------------------------------------------- # Compile # -------------------------------------------------------------------------------- learning_rate = 1e-3 if USE_SPARCE_LABEL: loss_fn=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False) else: loss_fn=tf.keras.losses.CategoricalCrossentropy(from_logits=False) model.compile( optimizer=Adam(learning_rate=learning_rate, beta_1=0.9, beta_2=0.999, epsilon=1e-08), loss=loss_fn, metrics=["accuracy"] ) model.summary() # -------------------------------------------------------------------------------- # Train # -------------------------------------------------------------------------------- batch_size = 16 number_of_epochs = 10 def data_label_generator(x, y): def _f(): index = 0 length = len(x) try: while True: yield x[index:index+batch_size], y[index:index+batch_size] index = (index + batch_size) % length except StopIteration: return return _f earlystop_callback = tf.keras.callbacks.EarlyStopping( patience=5, restore_best_weights=True, monitor='val_accuracy' ) steps_per_epoch = len(y_train) // batch_size validation_steps = (len(y_validation) // batch_size) - 1 # -1 to avoid run out of data for validation history = model.fit( x=data_label_generator(x_train, y_train)(), batch_size=batch_size, epochs=number_of_epochs, verbose=1, validation_data=data_label_generator(x_validation, y_validation)(), shuffle=True, steps_per_epoch=steps_per_epoch, validation_steps=validation_steps, validation_batch_size=batch_size, callbacks=[ earlystop_callback ] ) ``` ## Environment ``` TensorFlow version: 2.14.1 Keras version: 2.14.0 Python 3.10.12 Ubuntu 22.04LTS ``` [1]: https://stackoverflow.com/questions/64910527/accuracy-killed-when-using-imagedatagenerator-tensorflow-keras# ### Standalone code to reproduce the issue ```shell import numpy as np import tensorflow as tf from tensorflow import keras from keras import ( __version__ ) from keras.layers import ( Layer, Normalization, Conv2D, MaxPooling2D, BatchNormalization, Dense, Flatten, Dropout, Reshape, Activation, ReLU, LeakyReLU, ) from keras.models import ( Model, ) from keras.layers import ( Layer ) from keras.optimizers import ( Adam ) from sklearn.model_selection import train_test_split print("TensorFlow version: {}".format(tf.__version__)) tf.keras.__version__ = __version__ print("Keras version: {}".format(tf.keras.__version__)) # -------------------------------------------------------------------------------- # CIFAR 10 # -------------------------------------------------------------------------------- NUM_CLASSES = 10 INPUT_SHAPE = (32, 32, 3) USE_SPARCE_LABEL = False # Setting False make it work as expected (x_train, y_train), (x_test, y_test) = tf.keras.datasets.cifar10.load_data() x_train, x_test = x_train / 255.0, x_test / 255.0 x_train, x_validation, y_train, y_validation = train_test_split( x_train, y_train, test_size=0.2, random_state=42 ) # One Hot Encoding the labels if not USE_SPARCE_LABEL: y_train = keras.utils.to_categorical(y_train, NUM_CLASSES) y_validation = keras.utils.to_categorical(y_validation, NUM_CLASSES) y_test = keras.utils.to_categorical(y_test, NUM_CLASSES) # -------------------------------------------------------------------------------- # Model # -------------------------------------------------------------------------------- inputs = tf.keras.Input( name='image', shape=INPUT_SHAPE, dtype=tf.float32 ) x = Conv2D( filters=32, kernel_size=(3, 3), strides=(1, 1), padding="same", activation='relu', input_shape=INPUT_SHAPE )(inputs) x = BatchNormalization()(x) x = Conv2D( filters=64, kernel_size=(3, 3), strides=(1, 1), padding="same", activation='relu' )(x) x = MaxPooling2D( pool_size=(2, 2) )(x) x = Dropout(0.20)(x) x = Conv2D( filters=128, kernel_size=(3, 3), strides=(1, 1), padding="same", activation='relu' )(x) x = BatchNormalization()(x) x = MaxPooling2D( pool_size=(2, 2) )(x) x = Dropout(0.20)(x) x = Flatten()(x) x = Dense(300, activation="relu")(x) x = BatchNormalization()(x) x = Dropout(0.20)(x) x = Dense(200, activation="relu")(x) outputs = Dense(NUM_CLASSES, activation="softmax")(x) model: Model = Model( inputs=inputs, outputs=outputs, name="cifar10" ) # -------------------------------------------------------------------------------- # Compile # -------------------------------------------------------------------------------- learning_rate = 1e-3 if USE_SPARCE_LABEL: loss_fn=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=False) else: loss_fn=tf.keras.losses.CategoricalCrossentropy(from_logits=False) model.compile( optimizer=Adam(learning_rate=learning_rate, beta_1=0.9, beta_2=0.999, epsilon=1e-08), loss=loss_fn, metrics=["accuracy"] ) model.summary() # -------------------------------------------------------------------------------- # Train # -------------------------------------------------------------------------------- batch_size = 16 number_of_epochs = 10 def data_label_generator(x, y): def _f(): index = 0 length = len(x) try: while True: yield x[index:index+batch_size], y[index:index+batch_size] index = (index + batch_size) % length except StopIteration: return return _f earlystop_callback = tf.keras.callbacks.EarlyStopping( patience=5, restore_best_weights=True, monitor='val_accuracy' ) steps_per_epoch = len(y_train) // batch_size validation_steps = (len(y_validation) // batch_size) - 1 # -1 to avoid run out of data for validation history = model.fit( x=data_label_generator(x_train, y_train)(), batch_size=batch_size, epochs=number_of_epochs, verbose=1, validation_data=data_label_generator(x_validation, y_validation)(), shuffle=True, steps_per_epoch=steps_per_epoch, validation_steps=validation_steps, validation_batch_size=batch_size, callbacks=[ earlystop_callback ] ) ``` ### Relevant log output ```shell ## Symptom Using Sparse Index as the labels with and as the loss function (```USE_SPARSE_LABEL=True```). The accuracy values got unstable and low, causing early stop. ``` 2500/2500 [...] - 24s 8ms/step - loss: 1.4824 - accuracy: 0.0998 - val_loss: 1.1893 - val_accuracy: 0.1003 Epoch 2/10 2500/2500 [...] - 21s 8ms/step - loss: 1.0730 - accuracy: 0.1010 - val_loss: 0.8896 - val_accuracy: 0.0832 Epoch 3/10 2500/2500 [...] - 20s 8ms/step - loss: 0.9272 - accuracy: 0.1016 - val_loss: 0.9150 - val_accuracy: 0.0720 Epoch 4/10 2500/2500 [...] - 20s 8ms/step - loss: 0.7987 - accuracy: 0.1019 - val_loss: 0.8087 - val_accuracy: 0.0864 Epoch 5/10 2500/2500 [...] - 20s 8ms/step - loss: 0.7081 - accuracy: 0.1012 - val_loss: 0.8707 - val_accuracy: 0.0928 Epoch 6/10 2500/2500 [...] - 21s 8ms/step - loss: 0.6056 - accuracy: 0.1019 - val_loss: 0.7688 - val_accuracy: 0.0851 ``` Using One Hot Encoding as the labels and ```CategoricalCrossentropy``` as the loss function (```USE_SPARSE_LABEL=True```). Work as expected. ``` 2500/2500 [...] - 24s 8ms/step - loss: 1.4146 - accuracy: 0.4997 - val_loss: 1.0906 - val_accuracy: 0.6105 Epoch 2/10 2500/2500 [...] - 21s 9ms/step - loss: 1.0306 - accuracy: 0.6375 - val_loss: 0.9779 - val_accuracy: 0.6532 Epoch 3/10 2500/2500 [...] - 22s 9ms/step - loss: 0.8780 - accuracy: 0.6925 - val_loss: 0.8194 - val_accuracy: 0.7127 Epoch 4/10 2500/2500 [...] - 21s 8ms/step - loss: 0.7641 - accuracy: 0.7315 - val_loss: 0.9330 - val_accuracy: 0.7014 Epoch 5/10 2500/2500 [...] - 21s 8ms/step - loss: 0.6797 - accuracy: 0.7614 - val_loss: 0.7908 - val_accuracy: 0.7311 Epoch 6/10 2500/2500 [...] - 21s 9ms/step - loss: 0.6182 - accuracy: 0.7841 - val_loss: 0.7371 - val_accuracy: 0.7533 Epoch 7/10 2500/2500 [...] - 21s 9ms/step - loss: 0.4981 - accuracy: 0.8217 - val_loss: 0.8221 - val_accuracy: 0.7373 Epoch 8/10 2500/2500 [...] - 22s 9ms/step - loss: 0.4363 - accuracy: 0.8437 - val_loss: 0.7865 - val_accuracy: 0.7525 Epoch 9/10 2500/2500 [...] - 23s 9ms/step - loss: 0.3962 - accuracy: 0.8596 - val_loss: 0.8198 - val_accuracy: 0.7505 Epoch 10/10 2500/2500 [...] - 22s 9ms/step - loss: 0.3463 - accuracy: 0.8776 - val_loss: 0.8472 - val_accuracy: 0.7512 ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62460/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62460/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62459	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62459/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62459/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62459/events	https://github.com/tensorflow/tensorflow/issues/62459	2,006,951,489	I_kwDOArmXAs53n6ZB	62,459	Failure to compile TF 2.15.0/2.16.1 (with Cuda support) using clang in Ubuntu 22.04	{ "login": "feranick", "id": 35967, "node_id": "MDQ6VXNlcjM1OTY3", "avatar_url": "https://avatars.githubusercontent.com/u/35967?v=4", "gravatar_id": "", "url": "https://api.github.com/users/feranick", "html_url": "https://github.com/feranick", "followers_url": "https://api.github.com/users/feranick/followers", "following_url": "https://api.github.com/users/feranick/following{/other_user}", "gists_url": "https://api.github.com/users/feranick/gists{/gist_id}", "starred_url": "https://api.github.com/users/feranick/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/feranick/subscriptions", "organizations_url": "https://api.github.com/users/feranick/orgs", "repos_url": "https://api.github.com/users/feranick/repos", "events_url": "https://api.github.com/users/feranick/events{/privacy}", "received_events_url": "https://api.github.com/users/feranick/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 1205615612, "node_id": "MDU6TGFiZWwxMjA1NjE1NjEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:%20ubuntu/linux", "name": "subtype: ubuntu/linux", "color": "b619ea", "default": false, "description": "Ubuntu/Linux Build/Installation Issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "When nvcc is used as compiler, compilation still fails. Configure and compile Logs are attached.\r\n[nvcc_compile.txt](https://github.com/tensorflow/tensorflow/files/13443721/nvcc_compile.txt)\r\n[nvcc_config.txt](https://github.com/tensorflow/tensorflow/files/13443722/nvcc_config.txt)\r\n", "The only way to have a successful compilation is to use gcc with NO Cuda support.", "The original bug while using clang can be fixed by making sure the library `libstdc++-12-dev` is installed:\r\n\r\n`sudo apt install libstdc++-12-dev`\r\n\r\nHowever a new set of errors appear, which seem related to a confusing definition of `noinline`. Log:\r\n\r\n```\r\nERROR: /home/nicola/Software/tensorflow-dir/gpu/tensorflow/tensorflow/core/kernels/BUILD:5142:18: Compiling tensorflow/core/kernels/sparse_to_dense_op_gpu.cu.cc failed: (Exit 1): clang failed: error executing command (from target //tensorflow/core/kernels:sparse_to_dense_op_gpu) /usr/lib/llvm-14/bin/clang -MD -MF bazel-out/k8-opt/bin/tensorflow/core/kernels/_objs/sparse_to_dense_op_gpu/sparse_to_dense_op_gpu.cu.pic.d ... (remaining 221 arguments skipped)\r\nclang: warning: CUDA version is newer than the latest supported version 11.5 [-Wunknown-cuda-version]\r\nIn file included from tensorflow/core/kernels/sparse_to_dense_op_gpu.cu.cc:19:\r\nIn file included from ./tensorflow/core/kernels/sparse_to_dense_op_gpu.h:23:\r\nIn file included from external/local_xla/xla/stream_executor/device_memory.h:29:\r\nIn file included from external/local_xla/xla/stream_executor/platform/port.h:22:\r\nIn file included from external/local_tsl/tsl/platform/types.h:24:\r\nIn file included from external/local_tsl/tsl/platform/tstring.h:24:\r\nIn file included from external/local_tsl/tsl/platform/cord.h:21:\r\nIn file included from external/com_google_absl/absl/strings/cord.h:74:\r\nIn file included from external/com_google_absl/absl/base/internal/endian.h:22:\r\nIn file included from external/com_google_absl/absl/base/casts.h:28:\r\nIn file included from /usr/lib/gcc/x86_64-linux-gnu/12/../../../../include/c++/12/memory:76:\r\nIn file included from /usr/lib/gcc/x86_64-linux-gnu/12/../../../../include/c++/12/bits/shared_ptr.h:53:\r\n/usr/lib/gcc/x86_64-linux-gnu/12/../../../../include/c++/12/bits/shared_ptr_base.h:196:22: error: use of undeclared identifier 'noinline'; did you mean 'inline'?\r\n __attribute__((__noinline__))\r\n ^\r\nbazel-out/k8-opt/bin/external/local_config_cuda/cuda/cuda/include/crt/host_defines.h:83:24: note: expanded from macro '__noinline__'\r\n __attribute__((noinline))\r\n ^\r\nIn file included from tensorflow/core/kernels/sparse_to_dense_op_gpu.cu.cc:19:\r\nIn file included from ./tensorflow/core/kernels/sparse_to_dense_op_gpu.h:23:\r\nIn file included from external/local_xla/xla/stream_executor/device_memory.h:29:\r\nIn file included from external/local_xla/xla/stream_executor/platform/port.h:22:\r\nIn file included from external/local_tsl/tsl/platform/types.h:24:\r\nIn file included from external/local_tsl/tsl/platform/tstring.h:24:\r\nIn file included from external/local_tsl/tsl/platform/cord.h:21:\r\nIn file included from external/com_google_absl/absl/strings/cord.h:74:\r\nIn file included from external/com_google_absl/absl/base/internal/endian.h:22:\r\nIn file included from external/com_google_absl/absl/base/casts.h:28:\r\nIn file included from /usr/lib/gcc/x86_64-linux-gnu/12/../../../../include/c++/12/memory:76:\r\nIn file included from /usr/lib/gcc/x86_64-linux-gnu/12/../../../../include/c++/12/bits/shared_ptr.h:53:\r\n/usr/lib/gcc/x86_64-linux-gnu/12/../../../../include/c++/12/bits/shared_ptr_base.h:196:22: error: type name does not allow function specifier to be specified\r\nbazel-out/k8-opt/bin/external/local_config_cuda/cuda/cuda/include/crt/host_defines.h:83:24: note: expanded from macro '__noinline__'\r\n __attribute__((noinline))\r\n ^\r\nIn file included from tensorflow/core/kernels/sparse_to_dense_op_gpu.cu.cc:19:\r\nIn file included from ./tensorflow/core/kernels/sparse_to_dense_op_gpu.h:23:\r\nIn file included from external/local_xla/xla/stream_executor/device_memory.h:29:\r\nIn file included from external/local_xla/xla/stream_executor/platform/port.h:22:\r\nIn file included from external/local_tsl/tsl/platform/types.h:24:\r\nIn file included from external/local_tsl/tsl/platform/tstring.h:24:\r\nIn file included from external/local_tsl/tsl/platform/cord.h:21:\r\nIn file included from external/com_google_absl/absl/strings/cord.h:74:\r\nIn file included from external/com_google_absl/absl/base/internal/endian.h:22:\r\nIn file included from external/com_google_absl/absl/base/casts.h:28:\r\nIn file included from /usr/lib/gcc/x86_64-linux-gnu/12/../../../../include/c++/12/memory:76:\r\nIn file included from /usr/lib/gcc/x86_64-linux-gnu/12/../../../../include/c++/12/bits/shared_ptr.h:53:\r\n/usr/lib/gcc/x86_64-linux-gnu/12/../../../../include/c++/12/bits/shared_ptr_base.h:196:22: error: expected expression\r\nbazel-out/k8-opt/bin/external/local_config_cuda/cuda/cuda/include/crt/host_defines.h:83:33: note: expanded from macro '__noinline__'\r\n __attribute__((noinline))\r\n ^\r\n```", "@feranick Could you please try to uninstall the clang version 14 and install the clang 13 and try to reconfigure TF ? Please let us know if it helps?\r\nThank you!", "Compilation still fails, although for a different reason. Log attached.\r\n[clang-13.txt](https://github.com/tensorflow/tensorflow/files/13479822/clang-13.txt)\r\n", "Actually, I tried it in a different machine with a similar configuration, and even using clang-13 the error seems to be the same as in the original post (using clang-14). Log attached.\r\n\r\n[clang-13_2.txt](https://github.com/tensorflow/tensorflow/files/13480375/clang-13_2.txt)\r\n", "As the issue seems to be related to CUDA, what is the preferred version of CUDA that Google advises to use for TF2.15.0? TF2.14.x works with CUDA 11.8, but this doesn't work (as per this bug) for TF2.15.0.", "> As the issue seems to be related to CUDA, what is the preferred version of CUDA that Google advises to use for TF2.15.0? TF2.14.x works with CUDA 11.8, but this doesn't work (as per this bug) for TF2.15.0.\r\n\r\n12.2 is the CUDA version we bumped from TensorFlow 2.15. Could you please try with that and let us know the outcome. Thanks!", "Still getting the same error with CUDA 12.2. Log attached.\r\n[clang_3.txt](https://github.com/tensorflow/tensorflow/files/13490098/clang_3.txt)\r\n", "I just tried to compile TF2.15.0 using CUDA 12.2 and `nvcc` (NOT clang) and compilation was successful. I'd leave this bug report open, as compilation with clang still fails (and clang is now the default compiler in `configure`).\r\n", "It would also be good to update the ./configure to reflect that the min version of CUDA should be 12.2 (it still lists 11), and somewhere a list of the supported versions of each software required. ", "We have updated the document in the code, due to some sync issue it is not published yet in the website. \r\n", "I have the same problem on rocky 8.8 compiling tensorflow 2.15 with cuda 12.2 and clang 16.0.1. \r\n\r\nModules loaded for compilation (all compiled in house on the same host):\r\n\r\n```\r\nbazel/6.1.0 cuda/12.2.0 python/3.10.2 tensorRT/8.6.1.6 java/11 gcc/11.2.0 llvm/16.0.1 clang/16.0.1\r\n```\r\n\r\nRunning .configure results in the following .tf_configure.bazelrc (implying using clang as cuda compiler):\r\n\r\n```\r\nbuild --action_env PYTHON_BIN_PATH=\"/opt/apps/python/3.10.2/bin/python3\"\r\nbuild --action_env PYTHON_LIB_PATH=\"/opt/apps/python/3.10.2/lib/python3.10/site-packages\"\r\nbuild --python_path=\"/opt/apps/python/3.10.2/bin/python3\"\r\nbuild --config=tensorrt\r\nbuild --action_env TF_CUDA_VERSION=\"12\"\r\nbuild --action_env TF_CUDNN_VERSION=\"8\"\r\nbuild --action_env TF_TENSORRT_VERSION=\"8\"\r\nbuild --action_env TF_NCCL_VERSION=\"\"\r\nbuild --action_env TF_CUDA_PATHS=\"/opt/apps/cuda/12.2.0,/opt/apps/tensorRT/8.6.1.6,/usr\"\r\nbuild --action_env CUDA_TOOLKIT_PATH=\"/opt/apps/cuda/12.2.0\"\r\nbuild --action_env TF_CUDA_COMPUTE_CAPABILITIES=\"7.0,8.0\"\r\nbuild --action_env LD_LIBRARY_PATH=\"/opt/apps/clang/16.0.1/lib:/opt/apps/llvm/16.0.1/lib:/opt/apps/gcc/11.2.0/lib64:/opt/apps/gcc/11.2.0/lib:/opt/apps/gcc/11.2.0/lib/gcc/x86_64-pc-linux-gnu/11.2.0:/opt/apps/tensorRT/8.6.1.6/lib:/opt/apps/python/3.10.2\r\n/lib:/opt/apps/cuda/12.2.0/lib64:/opt/apps/cuda/12.2.0/nvvm/lib64:/opt/apps/cuda/12.2.0/cublas/lib64:/opt/apps/cuda/12.2.0/extras/CUPTI/lib64:/opt/apps/cuda/12.2.0/extras/Debugger/lib64\"\r\nbuild --config=cuda_clang\r\nbuild --action_env CLANG_CUDA_COMPILER_PATH=\"/opt/apps/clang/16.0.1/bin/clang-16\"\r\nbuild --copt=-Wno-gnu-offsetof-extensions\r\nbuild --config=cuda_clang\r\nbuild:opt --copt=-mavx2\r\nbuild:opt --host_copt=-mavx2\r\ntest --test_size_filters=small,medium\r\ntest --test_env=LD_LIBRARY_PATH\r\ntest:v1 --test_tag_filters=-benchmark-test,-no_oss,-oss_excluded,-gpu,-oss_serial\r\ntest:v1 --build_tag_filters=-benchmark-test,-no_oss,-oss_excluded,-gpu\r\ntest:v2 --test_tag_filters=-benchmark-test,-no_oss,-oss_excluded,-gpu,-oss_serial,-v1only\r\ntest:v2 --build_tag_filters=-benchmark-test,-no_oss,-oss_excluded,-gpu,-v1only\r\n```\r\n\r\nThe bazel build command:\r\n```\r\nbazel build --config=opt --jobs=8 --verbose_failures --verbose_explanations --explain=/tmp/explain \\\r\n //tensorflow/tools/pip_package:build_pip_package \r\n```\r\nIf i rerun configure and choose not to use clang as cuda compiler it results in a slightly \r\ndifferent .tf.configure.bazelrc file for brevity here are the diffs:\r\n\r\n```\r\ndont use clang\r\n< build --action_env GCC_HOST_COMPILER_PATH=\"/opt/apps/gcc/11.2.0/bin/gcc\"\r\n< build --config=cuda\r\n---\r\nuse clang\r\n> build --config=cuda_clang\r\n> build --action_env CLANG_CUDA_COMPILER_PATH=\"/opt/apps/clang/16.0.1/bin/clang-16\"\r\n> build --copt=-Wno-gnu-offsetof-extensions\r\n> build --config=cuda_clang\r\n```\r\n\r\nand the bazel build fails with a different error:\r\n\r\n```\r\nbazel-out/k8-opt-exec-50AE0418/bin/tensorflow/python/framework/offset_counter: /lib64/libstdc++.so.6: version 'CXXABI_1.3.13' not found (required by /root/.cache/bazel/_bazel_root/e76370378c3e9e8e238b869c10fc760e/execroot/org_tensorflow/bazel-out/k8-opt-exec-50AE0418/bin/tensorflow/python/framework/../../../_solib_local/_U_S_Stensorflow_Spython_Sframework_Coffset_Ucounter___Utensorflow/libtensorflow_framework.so.2)\r\nbazel-out/k8-opt-exec-50AE0418/bin/tensorflow/python/framework/offset_counter: /lib64/libstdc++.so.6: version 'GLIBCXX_3.4.29' not found (required by /root/.cache/bazel/_bazel_root/e76370378c3e9e8e238b869c10fc760e/execroot/org_tensorflow/bazel-out/k8-opt-exec-50AE0418/bin/tensorflow/python/framework/../../../_solib_local/_U_S_Stensorflow_Spython_Sframework_Coffset_Ucounter___Utensorflow/libtensorflow_framework.so.2)\r\nbazel-out/k8-opt-exec-50AE0418/bin/tensorflow/python/framework/offset_counter: /lib64/libstdc++.so.6: version 'GLIBCXX_3.4.26' not found (required by /root/.cache/bazel/_bazel_root/e76370378c3e9e8e238b869c10fc760e/execroot/org_tensorflow/bazel-out/k8-opt-exec-50AE0418/bin/tensorflow/python/framework/../../../_solib_local/_U_S_Stensorflow_Spython_Sframework_Coffset_Ucounter___Utensorflow/libtensorflow_framework.so.2)\r\n```\r\n\r\nWhile the compilation itself apparently went further, it makes no sense to me that bazel subcommand failed with \r\nabove errors because /lib64/libstdc++.so.6 should not be used as my LD_LIBRARY_PATH is set to use clang 16 \r\nand gcc 11.2.0 path first and gcc compilation includes libstdc++.so.6 that has all above mentioned CXX and GLIBCXX entries. Bazel binary and clang libraries are linked to use that correct libstdc++.so.6 from gcc 11. \r\nThe question is where does the /lib64/libstdc++.so.6 come from in bazel work? \r\n\r\nEither way, with or without clang bazel build fails. \r\n", "This is a separate issues than the one reported here (which is specific to Ubuntu, and related to the way CUDA is called within clang). I would recommend filing a separate issue for Rocky linux and libstfc++.", "I have also been trying to compile TF 2.15 (well, master actually) on Ubuntu 22.04 and have been running into some issues with clang.\r\n\r\nI have to say, I find the available documentation on compiling TF from source a tad contradictory and a tad confusing. My story so far is as follows.\r\n\r\n1) Install cuda 12.2 and clang 17 (as per 2.15 release notes).\r\n2) Install cuDnn 8.9.7.\r\n\r\nI then tried to compile master, but it fails because, despite not configuring tensorrt, it is looking for tensorrt headers.\r\n\r\n3) So I try to install tensorrt, but it does not support cuda 12.2 yet.\r\n4) Install cuda 12.1\r\n5) Install tensorrt\r\n6) Install libstdc++-12-dev \r\n\r\nCompiling with clang gets me deep into the compilation, but it then bombs out:\r\n\r\n```\r\nERROR: /home/christopher/.cache/bazel/_bazel_christopher/696f0d61a4ca964b6d054bb99f460f18/external/upb/BUILD:57:11: Compiling upb/decode.c failed: (Exit 1): clang failed: error executing command (from target @upb//:upb) /usr/lib/llvm-17/bin/clang -MD -MF bazel-out/k8-opt/bin/external/upb/_objs/upb/decode.pic.d '-frandom-seed=bazel-out/k8-opt/bin/external/upb/_objs/upb/decode.pic.o' '-DBAZEL_CURRENT_REPOSITORY=\"upb\"' ... (remaining 43 arguments skipped)\r\nclang: error: argument unused during compilation: '--cuda-path=/usr/local/cuda-12.1' [-Werror,-Wunused-command-line-argument]\r\nTarget //tensorflow/tools/pip_package:build_pip_package failed to build\r\n```\r\n\r\nI am now attempting the build with gcc ... ", "Still an issue with TF 2.1.6.1. Apparently compiling with clang supports only CUDA up to v11.5. See log below.\r\n[log.txt](https://github.com/tensorflow/tensorflow/files/14575336/log.txt)\r\n", "Have you tried it with Clang 17 on your ubuntu, here is the doc for installing Clang https://www.tensorflow.org/install/source#install_clang_recommended_linux_only", "I will. However, the updated clang 17 is not available in the standard Ubuntu repository (it's an external repo), which means that TF cannot be compiled wth Clang with standard tools. gcc works fine. It would be great to mention that during `./configure` (at least the min version of clang required. ", "I was able to build 2.16.1 with CUDA 12.4 with clang 17 on Ubuntu 22.04 by manually adding the line \"build:cuda --copt=-Wno-error=unused-command-line-argument\" into .tf_configure.bazelrc after running configure and before running bazel build. I also had to \"export TF_PYTHON_VERSION=3.10\" before \"bazel build --subcommands //tensorflow/tools/pip_package/v2:wheel --repo_env=WHEEL_NAME=tensorflow --config=cuda\". I do have a separate build error (see issue 62047) if I add copt \"-march=native\" or \"-mavx\" though.", "Minor correction - the issue referenced in 62047 only happens with \"-march=native\". \"-mavx\" does build and run successfully. The other issue was blocked from comments since I added my note there. This was done with Intel 4410Y CPUs.", "> I was able to build 2.16.1 with CUDA 12.4 with clang 17 on Ubuntu 22.04 by manually adding the line \"build:cuda --copt=-Wno-error=unused-command-line-argument\" into .tf_configure.bazelrc after running configure and before running bazel build. I also had to \"export TF_PYTHON_VERSION=3.10\" before \"bazel build --subcommands //tensorflow/tools/pip_package/v2:wheel --repo_env=WHEEL_NAME=tensorflow --config=cuda\". I do have a separate build error (see issue 62047) if I add copt \"-march=native\" or \"-mavx\" though.\r\n\r\nYou don't have to modify the `.tf_configure.bazelrc` file, but yes, if compiling with clang-17, then you should pass `--copt=-Wno-error=unused-command-line-argument` to the `bazel build` command and it will accomplish the same goal w/o modifying a file etc.\r\n", "I built v2.16.1 successfully with CUDA 12.3 + cuDNN 8.9.7 + tensorrt 8.6.1 with clang 17 on Ubuntu 22.04, with command line option `--copt=-Wno-error=unused-command-line-argument`." ]	2023-11-22T18:58:51	2024-04-09T00:47:05	null	CONTRIBUTOR	null	null	null	### Issue type Build/Install ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version 2.15.0/2.16.1 ### Custom code No ### OS platform and distribution Linux Ubuntu 16.04 ### Mobile device _No response_ ### Python version 3.10.12 ### Bazel version 6.1.0 ### GCC/compiler version Clang 14.0.0-1ubuntu1.1 ### CUDA/cuDNN version 11.8-12.3 ### GPU model and memory Quadro RTX 6000 24GB ### Current behavior? Compiling TF 2.15.0/2.16.1 in Ubuntu 22.04 using clang fails. Log is attached. ### Standalone code to reproduce the issue ```shell Using this configuration: ./configure You have bazel 6.1.0 installed. (6.5 for TF 2.16.1) Please specify the location of python. [Default is /usr/bin/python3]: Found possible Python library paths: /usr/lib/python3/dist-packages /usr/local/lib/python3.10/dist-packages Please input the desired Python library path to use. Default is [/usr/lib/python3/dist-packages] Do you wish to build TensorFlow with ROCm support? [y/N]: No ROCm support will be enabled for TensorFlow. Do you wish to build TensorFlow with CUDA support? [y/N]: y CUDA support will be enabled for TensorFlow. Do you wish to build TensorFlow with TensorRT support? [y/N]: y TensorRT support will be enabled for TensorFlow. Found CUDA 11.8 in: /usr/local/cuda-11.8/targets/x86_64-linux/lib /usr/local/cuda-11.8/targets/x86_64-linux/include Found cuDNN 8 in: /usr/lib/x86_64-linux-gnu /usr/include Found TensorRT 8.5.3 in: /usr/lib/x86_64-linux-gnu /usr/include/x86_64-linux-gnu Please specify a list of comma-separated CUDA compute capabilities you want to build with. You can find the compute capability of your device at: https://developer.nvidia.com/cuda-gpus. Each capability can be specified as "x.y" or "compute_xy" to include both virtual and binary GPU code, or as "sm_xy" to only include the binary code. Please note that each additional compute capability significantly increases your build time and binary size, and that TensorFlow only supports compute capabilities >= 3.5 [Default is: 7.5,7.5]: Do you want to use clang as CUDA compiler? [Y/n]: Clang will be used as CUDA compiler. Please specify clang path that to be used as host compiler. [Default is /usr/bin/clang]: You have Clang 14.0.0-1ubuntu1.1 installed. Please specify optimization flags to use during compilation when bazel option "--config=opt" is specified [Default is -Wno-sign-compare]: -Wno-sign-compare Would you like to interactively configure ./WORKSPACE for Android builds? [y/N]: Not configuring the WORKSPACE for Android builds. Preconfigured Bazel build configs. You can use any of the below by adding "--config=<>" to your build command. See .bazelrc for more details. --config=mkl # Build with MKL support. --config=mkl_aarch64 # Build with oneDNN and Compute Library for the Arm Architecture (ACL). --config=monolithic # Config for mostly static monolithic build. --config=numa # Build with NUMA support. --config=dynamic_kernels # (Experimental) Build kernels into separate shared objects. --config=v1 # Build with TensorFlow 1 API instead of TF 2 API. Preconfigured Bazel build configs to DISABLE default on features: --config=nogcp # Disable GCP support. --config=nonccl # Disable NVIDIA NCCL support. Configuration finished ``` ### Relevant log output ```shell Extracting Bazel installation... Starting local Bazel server and connecting to it... WARNING: The following configs were expanded more than once: [tensorrt, cuda_clang, cuda]. For repeatable flags, repeats are counted twice and may lead to unexpected behavior. INFO: Reading 'startup' options from /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --windows_enable_symlinks INFO: Options provided by the client: Inherited 'common' options: --isatty=1 --terminal_columns=100 INFO: Reading rc options for 'build' from /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: Inherited 'common' options: --experimental_repo_remote_exec INFO: Reading rc options for 'build' from /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: 'build' options: --define framework_shared_object=true --define tsl_protobuf_header_only=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=standalone -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --features=-force_no_whole_archive --enable_platform_specific_config --define=with_xla_support=true --config=short_logs --config=v2 --define=no_aws_support=true --define=no_hdfs_support=true --experimental_cc_shared_library --experimental_link_static_libraries_once=false --incompatible_enforce_config_setting_visibility INFO: Reading rc options for 'build' from /home/nicola/Software/tensorflow/gpu/tensorflow/.tf_configure.bazelrc: 'build' options: --action_env PYTHON_BIN_PATH=/usr/bin/python3 --action_env PYTHON_LIB_PATH=/usr/lib/python3/dist-packages --python_path=/usr/bin/python3 --config=tensorrt --action_env CUDA_TOOLKIT_PATH=/usr/local/cuda-11.8 --action_env TF_CUDA_COMPUTE_CAPABILITIES=7.5,7.5 --action_env LD_LIBRARY_PATH=/usr/local/cuda-11.8/include:/usr/local/cuda-11.8/targets/x86_64-linux/lib: --config=cuda_clang --action_env CLANG_CUDA_COMPILER_PATH=/usr/lib/llvm-14/bin/clang --config=cuda_clang INFO: Found applicable config definition build:short_logs in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --output_filter=DONT_MATCH_ANYTHING INFO: Found applicable config definition build:v2 in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1 INFO: Found applicable config definition build:tensorrt in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --repo_env TF_NEED_TENSORRT=1 INFO: Found applicable config definition build:cuda_clang in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --config=cuda --config=tensorrt --action_env=TF_CUDA_CLANG=1 --@local_config_cuda//:cuda_compiler=clang --repo_env=TF_CUDA_COMPUTE_CAPABILITIES=sm_50,sm_60,sm_70,sm_75,compute_80 INFO: Found applicable config definition build:cuda in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --repo_env TF_NEED_CUDA=1 --crosstool_top=@local_config_cuda//crosstool:toolchain --@local_config_cuda//:enable_cuda INFO: Found applicable config definition build:tensorrt in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --repo_env TF_NEED_TENSORRT=1 INFO: Found applicable config definition build:cuda_clang in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --config=cuda --config=tensorrt --action_env=TF_CUDA_CLANG=1 --@local_config_cuda//:cuda_compiler=clang --repo_env=TF_CUDA_COMPUTE_CAPABILITIES=sm_50,sm_60,sm_70,sm_75,compute_80 INFO: Found applicable config definition build:cuda in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --repo_env TF_NEED_CUDA=1 --crosstool_top=@local_config_cuda//crosstool:toolchain --@local_config_cuda//:enable_cuda INFO: Found applicable config definition build:tensorrt in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --repo_env TF_NEED_TENSORRT=1 INFO: Found applicable config definition build:opt in file /home/nicola/Software/tensorflow/gpu/tensorflow/.tf_configure.bazelrc: --copt=-Wno-sign-compare --host_copt=-Wno-sign-compare INFO: Found applicable config definition build:cuda in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --repo_env TF_NEED_CUDA=1 --crosstool_top=@local_config_cuda//crosstool:toolchain --@local_config_cuda//:enable_cuda INFO: Found applicable config definition build:linux in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --host_copt=-w --copt=-Wno-all --copt=-Wno-extra --copt=-Wno-deprecated --copt=-Wno-deprecated-declarations --copt=-Wno-ignored-attributes --copt=-Wno-array-bounds --copt=-Wunused-result --copt=-Werror=unused-result --copt=-Wswitch --copt=-Werror=switch --copt=-Wno-error=unused-but-set-variable --define=PREFIX=/usr --define=LIBDIR=$(PREFIX)/lib --define=INCLUDEDIR=$(PREFIX)/include --define=PROTOBUF_INCLUDE_PATH=$(PREFIX)/include --cxxopt=-std=c++17 --host_cxxopt=-std=c++17 --config=dynamic_kernels --experimental_guard_against_concurrent_changes INFO: Found applicable config definition build:dynamic_kernels in file /home/nicola/Software/tensorflow/gpu/tensorflow/.bazelrc: --define=dynamic_loaded_kernels=true --copt=-DAUTOLOAD_DYNAMIC_KERNELS DEBUG: /home/nicola/Software/tensorflow/gpu/tensorflow/tensorflow/tools/toolchains/python/python_repo.bzl:21:14: TF_PYTHON_VERSION variable was not set correctly, using default version. 3.10 Python will be used. To set Python version, run export TF_PYTHON_VERSION=3.9 WARNING: The following configs were expanded more than once: [tensorrt, cuda_clang, cuda]. For repeatable flags, repeats are counted twice and may lead to unexpected behavior. WARNING: Download from https://mirror.bazel.build/github.com/bazelbuild/rules_cc/archive/081771d4a0e9d7d3aa0eed2ef389fa4700dfb23e.tar.gz failed: class java.io.FileNotFoundException GET returned 404 Not Found INFO: Analyzed target //tensorflow/tools/pip_package:build_pip_package (699 packages loaded, 49725 targets configured). INFO: Found 1 target... ERROR: /home/nicola/.cache/bazel/_bazel_nicola/c53ed0be17816f9e0970b1ba234e403c/external/com_google_protobuf/BUILD.bazel:27:11: Compiling src/google/protobuf/arenaz_sampler.cc [for tool] failed: (Exit 1): clang failed: error executing command (from target @com_google_protobuf//:protobuf_lite) (cd /home/nicola/.cache/bazel/_bazel_nicola/c53ed0be17816f9e0970b1ba234e403c/execroot/org_tensorflow && \ exec env - \ LD_LIBRARY_PATH=/usr/local/cuda-11.8/include:/usr/local/cuda-11.8/targets/x86_64-linux/lib: \ PATH=/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin:/usr/games:/snap/bin \ PWD=/proc/self/cwd \ /usr/lib/llvm-14/bin/clang -MD -MF bazel-out/k8-opt-exec-50AE0418/bin/external/com_google_protobuf/_objs/protobuf_lite/arenaz_sampler.d '-frandom-seed=bazel-out/k8-opt-exec-50AE0418/bin/external/com_google_protobuf/_objs/protobuf_lite/arenaz_sampler.o' '-DBAZEL_CURRENT_REPOSITORY="com_google_protobuf"' -iquote external/com_google_protobuf -iquote bazel-out/k8-opt-exec-50AE0418/bin/external/com_google_protobuf -isystem external/com_google_protobuf/src -isystem bazel-out/k8-opt-exec-50AE0418/bin/external/com_google_protobuf/src -fmerge-all-constants -Wno-builtin-macro-redefined '-D__DATE__="redacted"' '-D__TIMESTAMP__="redacted"' '-D__TIME__="redacted"' -fPIE -U_FORTIFY_SOURCE '-D_FORTIFY_SOURCE=1' -fstack-protector -Wall -Wno-invalid-partial-specialization -fno-omit-frame-pointer -no-canonical-prefixes -DNDEBUG -g0 -O2 -ffunction-sections -fdata-sections '--cuda-path=/usr/local/cuda-11.8' -g0 -w -Wno-sign-compare -g0 '-std=c++17' -DHAVE_ZLIB -Woverloaded-virtual -Wno-sign-compare -c external/com_google_protobuf/src/google/protobuf/arenaz_sampler.cc -o bazel-out/k8-opt-exec-50AE0418/bin/external/com_google_protobuf/_objs/protobuf_lite/arenaz_sampler.o) # Configuration: fccf36a8de6040562b99997447208dc7348eb48df9f6a445d4378df7e13a876c # Execution platform: @local_execution_config_platform//:platform In file included from external/com_google_protobuf/src/google/protobuf/arenaz_sampler.cc:31: external/com_google_protobuf/src/google/protobuf/arenaz_sampler.h:34:10: fatal error: 'atomic' file not found #include <atomic> ^~~~~~~~ 1 error generated. Target //tensorflow/tools/pip_package:build_pip_package failed to build INFO: Elapsed time: 84.960s, Critical Path: 0.43s INFO: 60 processes: 59 internal, 1 local. FAILED: Build did NOT complete successfully ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62459/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62459/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62458	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62458/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62458/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62458/events	https://github.com/tensorflow/tensorflow/pull/62458	2,006,186,181	PR_kwDOArmXAs5gH2JT	62,458	build_pip_package: drop python_ppc64le files	{ "login": "smuzaffar", "id": 4115138, "node_id": "MDQ6VXNlcjQxMTUxMzg=", "avatar_url": "https://avatars.githubusercontent.com/u/4115138?v=4", "gravatar_id": "", "url": "https://api.github.com/users/smuzaffar", "html_url": "https://github.com/smuzaffar", "followers_url": "https://api.github.com/users/smuzaffar/followers", "following_url": "https://api.github.com/users/smuzaffar/following{/other_user}", "gists_url": "https://api.github.com/users/smuzaffar/gists{/gist_id}", "starred_url": "https://api.github.com/users/smuzaffar/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/smuzaffar/subscriptions", "organizations_url": "https://api.github.com/users/smuzaffar/orgs", "repos_url": "https://api.github.com/users/smuzaffar/repos", "events_url": "https://api.github.com/users/smuzaffar/events{/privacy}", "received_events_url": "https://api.github.com/users/smuzaffar/received_events", "type": "User", "site_admin": false }	[ { "id": 1169364259, "node_id": "MDU6TGFiZWwxMTY5MzY0MjU5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:XS", "name": "size:XS", "color": "adafea", "default": false, "description": "CL Change Size: Extra Small" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @jakeharmon8 Can you please review this PR ? Thank you!", "Hi @jakeharmon8 Can you please review this PR ? Thank you!", "Hi @jakeharmon8 Can you please review this PR ? Thank you!", "Hi @jakeharmon8 Can you please review this PR ? Thank you!", "Hi @smuzaffar Can you please resolve conflicts? Thank you!", "closing this as https://github.com/tensorflow/tensorflow/pull/64239 has removed this deprecated `tensorflow/tools/pip_package/build_pip_package.sh` file" ]	2023-11-22T11:57:55	2024-04-26T09:39:38	2024-04-26T09:39:32	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62458", "html_url": "https://github.com/tensorflow/tensorflow/pull/62458", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62458.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62458.patch", "merged_at": null }	Just like `python_x86_64` and `python_aarch64` , drop `python_ppc64le` files from the pip package	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62458/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62458/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62457	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62457/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62457/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62457/events	https://github.com/tensorflow/tensorflow/pull/62457	2,005,995,096	PR_kwDOArmXAs5gHMIE	62,457	[ppc64le] Add hh_vsx deps for highwayhash_dynamic	{ "login": "smuzaffar", "id": 4115138, "node_id": "MDQ6VXNlcjQxMTUxMzg=", "avatar_url": "https://avatars.githubusercontent.com/u/4115138?v=4", "gravatar_id": "", "url": "https://api.github.com/users/smuzaffar", "html_url": "https://github.com/smuzaffar", "followers_url": "https://api.github.com/users/smuzaffar/followers", "following_url": "https://api.github.com/users/smuzaffar/following{/other_user}", "gists_url": "https://api.github.com/users/smuzaffar/gists{/gist_id}", "starred_url": "https://api.github.com/users/smuzaffar/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/smuzaffar/subscriptions", "organizations_url": "https://api.github.com/users/smuzaffar/orgs", "repos_url": "https://api.github.com/users/smuzaffar/repos", "events_url": "https://api.github.com/users/smuzaffar/events{/privacy}", "received_events_url": "https://api.github.com/users/smuzaffar/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169364259, "node_id": "MDU6TGFiZWwxMTY5MzY0MjU5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:XS", "name": "size:XS", "color": "adafea", "default": false, "description": "CL Change Size: Extra Small" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[]	2023-11-22T10:09:59	2023-11-27T22:10:40	2023-11-27T22:10:40	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62457", "html_url": "https://github.com/tensorflow/tensorflow/pull/62457", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62457.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62457.patch", "merged_at": "2023-11-27T22:10:40" }	Building Tensorflow 2.15.0 for ppc64le failed with error like [a]. With this change I was able to fix this build issue. [a] ``` # Execution platform: @local_execution_config_platform//:platform ERROR: <path>/tensorflow-2.15.0/tensorflow/cc/BUILD:675:22: Linking tensorflow/cc/ops/user_ops_gen_cc [for tool] failed: (Exit 1): gcc failed: err ... ... # Execution platform: @local_execution_config_platform//:platform <gcc>/bin/ld: bazel-out/ppc-opt-exec-50AE0418/bin/_solib_ppc/_U_S_Stensorflow_Scc_Cops_Suser_Uops_Ugen_Ucc___Utensorflow/libtensorflow_framework.so.2: undefined reference to `highwayhash::HighwayHashCat<8u>::operator()(unsigned long const (&) [4], highwayhash::StringView const, unsigned long, unsigned long) const' <gcc>/bin/ld: bazel-out/ppc-opt-exec-50AE0418/bin/_solib_ppc/_U_S_Stensorflow_Scc_Cops_Suser_Uops_Ugen_Ucc___Utensorflow/libtensorflow_framework.so.2: undefined reference to `highwayhash::HighwayHash<8u>::operator()(unsigned long const (&) [4], char const, unsigned long, unsigned long) const' collect2: error: ld returned 1 exit status Target //tensorflow/tools/pip_package:build_pip_package failed to build ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62457/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62457/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62456	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62456/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62456/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62456/events	https://github.com/tensorflow/tensorflow/issues/62456	2,005,967,995	I_kwDOArmXAs53kKR7	62,456	Unclear exception when using resize_with_pad on images with original_image_ratio > target_side	{ "login": "EricPulido", "id": 47590491, "node_id": "MDQ6VXNlcjQ3NTkwNDkx", "avatar_url": "https://avatars.githubusercontent.com/u/47590491?v=4", "gravatar_id": "", "url": "https://api.github.com/users/EricPulido", "html_url": "https://github.com/EricPulido", "followers_url": "https://api.github.com/users/EricPulido/followers", "following_url": "https://api.github.com/users/EricPulido/following{/other_user}", "gists_url": "https://api.github.com/users/EricPulido/gists{/gist_id}", "starred_url": "https://api.github.com/users/EricPulido/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/EricPulido/subscriptions", "organizations_url": "https://api.github.com/users/EricPulido/orgs", "repos_url": "https://api.github.com/users/EricPulido/repos", "events_url": "https://api.github.com/users/EricPulido/events{/privacy}", "received_events_url": "https://api.github.com/users/EricPulido/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @EricPulido ,\r\n\r\nI have tested the code. The error is arising from C++ backend hence description is not clear.It seems generated from this line of code from ResizeBilinearOp.\r\n\r\nhttps://github.com/tensorflow/tensorflow/blob/2d7328762fa39668b1f802788f6143c4ffa1e1c4/tensorflow/core/kernels/image/resize_bilinear_op.cc#L58 \r\n\r\nwhich in turn calling output validation from here.\r\n\r\nhttps://github.com/tensorflow/tensorflow/blob/b487f0246c5804978f86f0a6bf2add1b911af23c/tensorflow/core/util/image_resizer_state.h#L123-L124\r\n\r\nHere given `'input_size'` is ( 1,2000). When it scales the input size using `'target_size'` (768,768), the `'input_size'` dimension becomes `'0'` making the above check fails and generating the error. You can refer the [gist](https://colab.sandbox.google.com/gist/SuryanarayanaY/0310f3f305f0e66e3d7d3f34b0690dfc/62456.ipynb) for aforesaid calculations.\r\n\r\n\r\n\r\n>When using tf.image.resize_with_pad, if the ratio of the dimensions of the original image is bigger than one of the transformations width or height, then it will fail with\r\noutput dimensions must be positive [Op:ResizeBilinear]\r\n\r\nI am not actually getting this. Can you please elaborate ?\r\n\r\nThanks!\r\n\r\n", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62456\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62456\">No</a>\n" ]	2023-11-22T09:56:11	2023-12-09T01:48:11	2023-12-09T01:48:08	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version v1.12.1-101328-g60ccf67fad1 2.16.0-dev20231018 ### Custom code Yes ### OS platform and distribution macOS 13.6.2 ### Mobile device _No response_ ### Python version 3.11 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? When using `tf.image.resize_with_pad`, if the ratio of the dimensions of the original image is bigger than one of the transformations width or height, then it will fail with ```output dimensions must be positive [Op:ResizeBilinear]``` While it's true that the process itself as set up generates output dimensions that are zero, this behaviour seems unexpected since it's possible to individually resize and pad without issue. ### Standalone code to reproduce the issue ```shell import tensorflow as tf image = tf.zeros(shape=(1, 2000, 3)) out = tf.image.resize_with_pad( image, target_height=768, target_width=768, ) ``` ``` ### Relevant log output ```shell Traceback (most recent call last): File "...venv/lib/python3.11/site-packages/tensorflow/python/util/traceback_utils.py", line 153, in error_handler raise e.with_traceback(filtered_tb) from None File "...venv/lib/python3.11/site-packages/tensorflow/python/framework/ops.py", line 5883, in raise_from_not_ok_status raise core._status_to_exception(e) from None # pylint: disable=protected-access ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ tensorflow.python.framework.errors_impl.InvalidArgumentError: {{function_node __wrapped__ResizeBilinear_device_/job:localhost/replica:0/task:0/device:CPU:0}} output dimensions must be positive [Op:ResizeBilinear] name: python-BaseException ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62456/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62456/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62455	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62455/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62455/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62455/events	https://github.com/tensorflow/tensorflow/issues/62455	2,005,904,328	I_kwDOArmXAs53j6vI	62,455	ByteBuffer is not a valid TensorFlow Lite model flatbuffe	{ "login": "yonaimineakio", "id": 43339576, "node_id": "MDQ6VXNlcjQzMzM5NTc2", "avatar_url": "https://avatars.githubusercontent.com/u/43339576?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yonaimineakio", "html_url": "https://github.com/yonaimineakio", "followers_url": "https://api.github.com/users/yonaimineakio/followers", "following_url": "https://api.github.com/users/yonaimineakio/following{/other_user}", "gists_url": "https://api.github.com/users/yonaimineakio/gists{/gist_id}", "starred_url": "https://api.github.com/users/yonaimineakio/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yonaimineakio/subscriptions", "organizations_url": "https://api.github.com/users/yonaimineakio/orgs", "repos_url": "https://api.github.com/users/yonaimineakio/repos", "events_url": "https://api.github.com/users/yonaimineakio/events{/privacy}", "received_events_url": "https://api.github.com/users/yonaimineakio/received_events", "type": "User", "site_admin": false }	[ { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1661751498, "node_id": "MDU6TGFiZWwxNjYxNzUxNDk4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TFLiteConverter", "name": "TFLiteConverter", "color": "bfdadc", "default": false, "description": "For issues related to TFLite converter" } ]	closed	false	{ "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false }	[ { "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false } ]	null	[ "Yo, not sure if you've done it already but I asked chatGPT your question by providing it all the code you've given here and the answer is below. I highly recommend to befriend chatGPT and make him your best friend. \r\nP.S, \r\nGPT is a native English speaker that speaks other language fluently, i.e. talk to him only in English. \r\n\r\nGPT>>>>\r\nThe error you're encountering suggests that the ByteBuffer you're trying to use as a TensorFlow Lite model is not valid. This could happen due to a few reasons:\r\n\r\nModel Conversion: Ensure the conversion from the SavedModel to the TFLite model was successful without any errors or warnings. Check if there were any warnings or issues during the conversion process in Google Colab.\r\n\r\nModel Loading: Verify the TFLite model file (onebar_model_div4.tflite) generated in Google Colab is being loaded correctly into your Android application. Ensure the file path and name are accurate.\r\n\r\nByteBuffer Mapping: Ensure that the ByteBuffer being created in your Android application accurately represents the TFLite model. The mapping of the ByteBuffer from the file should encompass the entire model size without any discrepancies.\r\n\r\nIn the provided Kotlin code, the function loadModelFile seems to be responsible for loading the TFLite model as a ByteBuffer. Check the values of fileDescriptor?.declaredLength and fileDescriptor?.startOffset to ensure they point to the right locations within the file for mapping.\r\n\r\nAdditionally, confirm that the model file is correctly packaged within the Android app's assets directory.\r\n\r\nIf everything seems accurate with the file and loading process, there might be an issue with the TFLite model conversion itself or with the way it's being read in the Android code.\r\n\r\nConsider adding some checks and logging statements in your Android code to print out relevant information like file paths, sizes, and offsets to ensure they match your expectations. This can help pinpoint where the issue might be occurring.\r\n\r\n", "@yonaimineakio Could you please check the size of your model file and if you copied the model file from another location, make sure that the file was copied correctly and that there are no errors in the file name or path so try to re-download it as well. Please make sure you are using the correct model format of tflite. Please let us know the TF version you are using?\r\nThank you!", "@sushreebarsa \r\nThank you for your response.\r\n\r\nmodel size: 32.53MB\r\ntf version: 2.14.0\r\n\r\nI copied another location without errors, Checked that file name and path was correct.\r\nI think I`m using correct model format of tflite because I coverted model following [official instructions](https://www.tensorflow.org/lite/convert/python_api?hl=ja).\r\n\r\ncodes is below.\r\n```\r\nimport tensorflow as tf\r\nfrom tensorflow import keras\r\nimport shutil\r\nimport numpy as np\r\n\r\nshutil.unpack_archive(\"/content/drive/MyDrive/試行錯誤/onebar_model_div4.zip\", \"onebar_model_div4\")\r\n\r\noutput_model = \"/content/drive/MyDrive/作業フォルダ/onebar_model_div4.tflite\"\r\nmodel = tf.keras.models.load_model(\"/content/onebar_model_div4/onebar_model_div4\")\r\nconverter = tf.lite.TFLiteConverter.from_keras_model(model)\r\ntflite_model = converter.convert()\r\n\r\nwith open(output_model, \"wb\") as t:\r\n t.write(tflite_model) \r\n```\r\n", "@TombCrawler \r\nThank you for your response.\r\nyes, I asked gpt4 but I thought the answears generated not directly related to the solution.", "![image](https://github.com/tensorflow/tensorflow/assets/43339576/0f1362a3-6ce3-41cc-9422-697660d0adee)\r\n\r\n", "![image](https://github.com/tensorflow/tensorflow/assets/43339576/483b121a-27c7-4026-b8b4-c785578c08a2)\r\n\r\n", "Sorry. I wroted the arguments incorrectly.\r\nI fixed code like below then the issue solved.\r\n\r\nBefor\r\n```\r\n @Throws(IOException::class)\r\n private fun loadModelFile(assets: AssetManager?, modelFileName: String): MappedByteBuffer{\r\n val fileDescriptor = assets?.openFd(modelFileName)\r\n val inputStream = FileInputStream(fileDescriptor?.fileDescriptor)\r\n return inputStream.channel.map(\r\n FileChannel.MapMode.READ_ONLY,\r\n fileDescriptor?.declaredLength ?: 0, \r\n fileDescriptor?.startOffset ?: 0 )\r\n}\r\n```\r\nAfter\r\n```\r\n @Throws(IOException::class)\r\n private fun loadModelFile(assets: AssetManager?, modelFileName: String): MappedByteBuffer{\r\n val fileDescriptor = assets?.openFd(modelFileName)\r\n val inputStream = FileInputStream(fileDescriptor?.fileDescriptor)\r\n return inputStream.channel.map(\r\n FileChannel.MapMode.READ_ONLY,\r\n fileDescriptor?.startOffset ?: 0 // fixed\r\n fileDescriptor?.declaredLength ?: 0,\r\n )\r\n}\r\n```", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62455\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62455\">No</a>\n" ]	2023-11-22T09:20:21	2023-12-04T06:04:52	2023-11-28T09:55:33	NONE	null	null	null	### 1. System information I`ve coverted an SavedMolde into TFlite model on GoogleColab then integrate it into Androi Application, building on Android Studio (bumblebee 2021.1.1 patch2) But when running application, get blow error. > >W/System.err: java.lang.IllegalArgumentException: ByteBuffer is not a valid TensorFlow Lite model flatbuffer > W/System.err: at org.tensorflow.lite.NativeInterpreterWrapper.createModelWithBuffer(Native Method) > W/System.err: at org.tensorflow.lite.NativeInterpreterWrapper.<init>(NativeInterpreterWrapper.java:72) > W/System.err: at org.tensorflow.lite.NativeInterpreterWrapperExperimental.<init>(NativeInterpreterWrapperExperimental.java:36) > W/System.err: at org.tensorflow.lite.Interpreter.<init>(Interpreter.java:232) > W/System.err: at org.tensorflow.lite.Interpreter.<init>(Interpreter.java:216) > W/System.err: at jp.kthrlab.jamsketch.JamSketchEngineTF.musicCalculatorForOutline(JamSketchEngineTF.kt:17) > W/System.err: at jp.kthrlab.jamsketch.JamSketchEngineAbstract.init(JamSketchEngineAbstract.kt:64) > W/System.err: at jp.kthrlab.jamsketch.MelodyData2.<init>(MelodyData2.kt:34) > W/System.err: at jp.kthrlab.jamsketch.JamSketch.initData(JamSketch.kt:104) > W/System.err: at jp.kthrlab.jamsketch.JamSketch.startMusic(JamSketch.kt:189) > W/System.err: at java.lang.reflect.Method.invoke(Native Method) > W/System.err: at processing.core.PApplet.method(PApplet.java:2905) > W/System.err: at jp.kthrlab.jamsketch.Button.mouseEvent(Button.java:43) > W/System.err: at java.lang.reflect.Method.invoke(Native Method) > W/System.err: at processing.core.PApplet$RegisteredMethods.handle(PApplet.java:1010) > W/System.err: at processing.core.PApplet$3.run(PApplet.java:1217) > W/System.err: at android.os.Handler.handleCallback(Handler.java:790) > W/System.err: at android.os.Handler.dispatchMessage(Handler.java:99) > W/System.err: at android.os.Looper.loop(Looper.java:169) > W/System.err: at android.app.ActivityThread.main(ActivityThread.java:6521) > W/System.err: at java.lang.reflect.Method.invoke(Native Method) > W/System.err: at com.android.internal.os.RuntimeInit$MethodAndArgsCaller.run(RuntimeInit.java:438) > W/System.err: at com.android.internal.os.ZygoteInit.main(ZygoteInit.java:807) > I/Choreographer: Skipped 118 frames! The application may be doing too much work on its main thread. > ### 2. Code colab notebook https://colab.research.google.com/drive/1HXVRiqTu4Yy7JdXSpja-rc77F0g084hU?usp=sharing Relevent kotlin file1. loading tflite model as byteBuffer type and assigned it to TF_MODEL. ``` val assetManager: AssetManager? = JamSketchActivity.myResources?.getAssets() var TF_MODEL: MappedByteBuffer? = null TF_MODEL = loadModelFile(assetManager, Config.TFL_MODEL_FILE) . . . . . . . @Throws(IOException::class) private fun loadModelFile(assets: AssetManager?, modelFileName: String): MappedByteBuffer{ val fileDescriptor = assets?.openFd(modelFileName) val inputStream = FileInputStream(fileDescriptor?.fileDescriptor) // println("declaredLength: ${fileDescriptor?.declaredLength}") // println("startoffset: ${fileDescriptor?.startOffset}") return inputStream.channel.map( FileChannel.MapMode.READ_ONLY, fileDescriptor?.declaredLength ?: 0, fileDescriptor?.startOffset ?: 0 ) } ``` Relevent kotlin file2. Wrapped the TF_MODEL with Interpreter then Issue have occured. ``` override fun musicCalculatorForOutline(): MusicCalculator { val noteSeqGenerator = NoteSeqGenerator(MELODY_LAYER, CHORD_LAYER, Config.BEATS_PER_MEASURE, Config.ENT_BIAS, model!!) tflite = Interpreter(TF_MODEL!!) printTensor(tflite!!) return noteSeqGenerator } ``` I referenced the similir [issue] (https://github.com/tensorflow/tensorflow/issues/44472) but I think it is not getting to the solution. I have been stucking this error over 1 week, so if anybody know tips getting to solution Please teach me. Please let me know if you need more information.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62455/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62455/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62454	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62454/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62454/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62454/events	https://github.com/tensorflow/tensorflow/issues/62454	2,005,755,325	I_kwDOArmXAs53jWW9	62,454	XLA for Accelerated Linear Algebra should be called Non-Linear Algebra	{ "login": "joekohler", "id": 147182672, "node_id": "U_kgDOCMXUUA", "avatar_url": "https://avatars.githubusercontent.com/u/147182672?v=4", "gravatar_id": "", "url": "https://api.github.com/users/joekohler", "html_url": "https://github.com/joekohler", "followers_url": "https://api.github.com/users/joekohler/followers", "following_url": "https://api.github.com/users/joekohler/following{/other_user}", "gists_url": "https://api.github.com/users/joekohler/gists{/gist_id}", "starred_url": "https://api.github.com/users/joekohler/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/joekohler/subscriptions", "organizations_url": "https://api.github.com/users/joekohler/orgs", "repos_url": "https://api.github.com/users/joekohler/repos", "events_url": "https://api.github.com/users/joekohler/events{/privacy}", "received_events_url": "https://api.github.com/users/joekohler/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1133285679, "node_id": "MDU6TGFiZWwxMTMzMjg1Njc5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:xla", "name": "comp:xla", "color": "0052cc", "default": false, "description": "XLA" } ]	closed	false	{ "login": "Venkat6871", "id": 147127861, "node_id": "U_kgDOCMT-NQ", "avatar_url": "https://avatars.githubusercontent.com/u/147127861?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Venkat6871", "html_url": "https://github.com/Venkat6871", "followers_url": "https://api.github.com/users/Venkat6871/followers", "following_url": "https://api.github.com/users/Venkat6871/following{/other_user}", "gists_url": "https://api.github.com/users/Venkat6871/gists{/gist_id}", "starred_url": "https://api.github.com/users/Venkat6871/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Venkat6871/subscriptions", "organizations_url": "https://api.github.com/users/Venkat6871/orgs", "repos_url": "https://api.github.com/users/Venkat6871/repos", "events_url": "https://api.github.com/users/Venkat6871/events{/privacy}", "received_events_url": "https://api.github.com/users/Venkat6871/received_events", "type": "User", "site_admin": false }	[ { "login": "Venkat6871", "id": 147127861, "node_id": "U_kgDOCMT-NQ", "avatar_url": "https://avatars.githubusercontent.com/u/147127861?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Venkat6871", "html_url": "https://github.com/Venkat6871", "followers_url": "https://api.github.com/users/Venkat6871/followers", "following_url": "https://api.github.com/users/Venkat6871/following{/other_user}", "gists_url": "https://api.github.com/users/Venkat6871/gists{/gist_id}", "starred_url": "https://api.github.com/users/Venkat6871/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Venkat6871/subscriptions", "organizations_url": "https://api.github.com/users/Venkat6871/orgs", "repos_url": "https://api.github.com/users/Venkat6871/repos", "events_url": "https://api.github.com/users/Venkat6871/events{/privacy}", "received_events_url": "https://api.github.com/users/Venkat6871/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @joekohler,\r\nWe see that the issue [template]( https://github.com/tensorflow/tensorflow/issues/new/choose) has not been filled, could you please do so as it helps us analyze the issue [tf version, steps followed before you ran into this error or stand alone code/colab gist to reproduce the issue faced.\r\n\r\nThank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62454\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62454\">No</a>\n" ]	2023-11-22T07:44:04	2024-01-03T01:48:45	2024-01-03T01:48:42	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2 ### Custom code Yes ### OS platform and distribution Android ### Mobile device Google Pixel ### Python version 3.14(dev) ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I expected exactly what I thought was going to happen, to happen ### Standalone code to reproduce the issue ```shell https://github.com/tensorflow/tensorflow/labels/good%20first%20issue ``` ### Relevant log output ```shell Tensorflow creating a good first issue ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62454/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62454/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62453	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62453/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62453/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62453/events	https://github.com/tensorflow/tensorflow/pull/62453	2,004,938,516	PR_kwDOArmXAs5gDm9P	62,453	[TOSA] add tests for legalization of quantized MUL,SUB,ADD operators with unequal ranks	{ "login": "psunn", "id": 55463253, "node_id": "MDQ6VXNlcjU1NDYzMjUz", "avatar_url": "https://avatars.githubusercontent.com/u/55463253?v=4", "gravatar_id": "", "url": "https://api.github.com/users/psunn", "html_url": "https://github.com/psunn", "followers_url": "https://api.github.com/users/psunn/followers", "following_url": "https://api.github.com/users/psunn/following{/other_user}", "gists_url": "https://api.github.com/users/psunn/gists{/gist_id}", "starred_url": "https://api.github.com/users/psunn/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/psunn/subscriptions", "organizations_url": "https://api.github.com/users/psunn/orgs", "repos_url": "https://api.github.com/users/psunn/repos", "events_url": "https://api.github.com/users/psunn/events{/privacy}", "received_events_url": "https://api.github.com/users/psunn/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 1169365494, "node_id": "MDU6TGFiZWwxMTY5MzY1NDk0", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:M", "name": "size:M", "color": "adafea", "default": false, "description": "CL Change Size: Medium" } ]	open	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "This PR is relative to https://github.com/tensorflow/tensorflow/pull/60753 and https://github.com/tensorflow/tensorflow/pull/62153", "Hi @jpienaar Can you please review this PR ? Thank you!", "Hi @jpienaar Can you please review this PR ? Thank you!", "Hi @jpienaar Can you please review this PR ? Thank you!", "Hi @jpienaar Can you please review this PR ? Thank you!", "Hi @jpienaar Can you please review this PR ? Thank you!", "Hi @jpienaar Can you please review this PR ? Thank you!" ]	2023-11-21T18:50:51	2024-06-07T16:36:04	null	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62453", "html_url": "https://github.com/tensorflow/tensorflow/pull/62453", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62453.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62453.patch", "merged_at": null }	add LIT tests for legalization of quantized MUL,SUB,ADD operators with unequal ranks	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62453/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62453/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62452	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62452/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62452/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62452/events	https://github.com/tensorflow/tensorflow/pull/62452	2,004,676,972	PR_kwDOArmXAs5gCtXN	62,452	[tosa] fix legalization of SquaredDifference op for unequal ranks	{ "login": "psunn", "id": 55463253, "node_id": "MDQ6VXNlcjU1NDYzMjUz", "avatar_url": "https://avatars.githubusercontent.com/u/55463253?v=4", "gravatar_id": "", "url": "https://api.github.com/users/psunn", "html_url": "https://github.com/psunn", "followers_url": "https://api.github.com/users/psunn/followers", "following_url": "https://api.github.com/users/psunn/following{/other_user}", "gists_url": "https://api.github.com/users/psunn/gists{/gist_id}", "starred_url": "https://api.github.com/users/psunn/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/psunn/subscriptions", "organizations_url": "https://api.github.com/users/psunn/orgs", "repos_url": "https://api.github.com/users/psunn/repos", "events_url": "https://api.github.com/users/psunn/events{/privacy}", "received_events_url": "https://api.github.com/users/psunn/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 1169364458, "node_id": "MDU6TGFiZWwxMTY5MzY0NDU4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:S", "name": "size:S", "color": "adafea", "default": false, "description": "CL Change Size: Small" } ]	open	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "This PR is relative to https://github.com/tensorflow/tensorflow/pull/60753", "Hi @rdzhabarov Can you please review this PR ? Thank you!", "Hi @rdzhabarov Can you please review this PR ? Thank you!", "Hi @rdzhabarov Can you please review this PR ? Thank you!", "Hi @rdzhabarov Can you please review this PR ? Thank you!" ]	2023-11-21T16:17:24	2024-06-07T16:35:42	null	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62452", "html_url": "https://github.com/tensorflow/tensorflow/pull/62452", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62452.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62452.patch", "merged_at": null }	This patch add EqualizeRanks of SquaredDifference Operator's inputs in case their ranks are different. LIT test is added to test for this case.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62452/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62452/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62451	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62451/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62451/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62451/events	https://github.com/tensorflow/tensorflow/issues/62451	2,004,495,593	I_kwDOArmXAs53eizp	62,451	`tf.linalg.cholesky` Produces Inconsistent Results with Complex and Float Tensors	{ "login": "Gwihwan-Go", "id": 83144588, "node_id": "MDQ6VXNlcjgzMTQ0NTg4", "avatar_url": "https://avatars.githubusercontent.com/u/83144588?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Gwihwan-Go", "html_url": "https://github.com/Gwihwan-Go", "followers_url": "https://api.github.com/users/Gwihwan-Go/followers", "following_url": "https://api.github.com/users/Gwihwan-Go/following{/other_user}", "gists_url": "https://api.github.com/users/Gwihwan-Go/gists{/gist_id}", "starred_url": "https://api.github.com/users/Gwihwan-Go/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Gwihwan-Go/subscriptions", "organizations_url": "https://api.github.com/users/Gwihwan-Go/orgs", "repos_url": "https://api.github.com/users/Gwihwan-Go/repos", "events_url": "https://api.github.com/users/Gwihwan-Go/events{/privacy}", "received_events_url": "https://api.github.com/users/Gwihwan-Go/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi,\r\n\r\nIt seems the issue might be related to the requirements for Cholesky decomposition. For the decomposition to be valid, especially with complex matrices, each input matrix needs to be Hermitian (a matrix that is equal to its own conjugate transpose) and positive definite (all its eigenvalues are positive).\r\n\r\nHere are some references for more details:\r\n\r\nhttps://en.wikipedia.org/wiki/Cholesky_decomposition\r\nhttps://www.tensorflow.org/api_docs/python/tf/linalg/cholesky\r\n\r\nAlso, here's a code snippet to check if your matrices meet these requirements:\r\n\r\n```python\r\nimport tensorflow as tf\r\nimport numpy as np\r\n\r\nparams = [\r\n tf.complex(tf.random.uniform([1, 26, 2, 2], dtype=tf.float32), tf.random.uniform([1, 26, 2, 2], dtype=tf.float32)),\r\n]\r\n\r\nclass Model1(tf.keras.Model):\r\n def __init__(self):\r\n super().__init__()\r\n self.p0 = tf.Variable(params[0]) # [1, 26, 2, 2] complex64\r\n\r\n @tf.function\r\n def __call__(self, v7_0):\r\n # Check if each matrix in the batch is Hermitian\r\n is_hermitian = tf.reduce_all(tf.math.equal(self.p0, tf.linalg.adjoint(self.p0)), axis=[-2, -1])\r\n\r\n # Calculate the eigenvalues for each matrix in the batch\r\n eigenvalues = tf.linalg.eigvalsh(self.p0)\r\n\r\n # Check for positive definiteness by ensuring the real parts of the eigenvalues are positive for each matrix\r\n is_positive_definite = tf.reduce_all(tf.math.real(eigenvalues) > 0, axis=-1)\r\n\r\n # Check if all matrices in the batch are Hermitian and positive definite\r\n valid_for_cholesky = tf.logical_and(is_hermitian, is_positive_definite)\r\n\r\n if tf.reduce_all(valid_for_cholesky):\r\n cho = tf.linalg.cholesky(self.p0)\r\n return cho \r\n else:\r\n print(\"is_hermitian: \", is_hermitian.numpy())\r\n print(\"is_positive_definite: \", is_positive_definite.numpy())\r\n raise ValueError(\"Not all matrices are Hermitian positive definite.\")\r\n\r\ninputs = [\r\n tf.complex(tf.random.uniform([1, 26, 1, 2], dtype=tf.float32), tf.random.uniform([1, 26, 1, 2], dtype=tf.float32)),\r\n]\r\n\r\nmodel1 = Model1()\r\n\r\nwith tf.device('cpu'):\r\n tf.config.run_functions_eagerly(True)\r\n out1 = model1(inputs)\r\n out2 = model1(inputs)\r\n print(f'=========eager_output(version:{tf.__version__})================')\r\n try:\r\n for i in range(len(out1)):\r\n np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.01, err_msg=f'at checking {i}th')\r\n print(\"Eager does not trigger assertion\")\r\n except AssertionError as e:\r\n print(\"Eager triggers assertion\")\r\n print(e)\r\n```", "Thanks for your response. I've revisited the [TensorFlow documentation on Cholesky](https://www.tensorflow.org/api_docs/python/tf/linalg/cholesky) and carefully checked the requirements:\r\n\r\nThe input must be a tensor with dimensions [..., M, M], where the last two dimensions are square matrices.\r\nThe tensor should be symmetric and positive definite. The function only uses the lower-triangular part of the tensor, ignoring the upper-triangular part.\r\nThe output is a tensor of the same shape, containing the Cholesky decompositions for all submatrices [..., :, :].\r\nConsidering these requirements, I updated my code to ensure it meets both criteria: forming square matrices and being symmetric and positive definite. Despite this, I'm still encountering an error. This leads me to think the issue might not be with my implementation but could potentially be an issue within TensorFlow's handling of the Cholesky operation. Here's the code I updated, which I believe adheres to the documented requirements:\r\n```python\r\nimport tensorflow as tf\r\nimport numpy as np\r\n\r\ndef generate_symmetric_positive_complex_tensor(shape):\r\n # Ensure the innermost shape is square\r\n assert shape[-1] == shape[-2], \"Shape must be square to create a symmetric matrix\"\r\n\r\n # Create a symmetric matrix for the real part and positive definite\r\n real_part = tf.random.uniform(shape, minval=0, maxval=1)\r\n real_symmetric = (real_part + tf.transpose(real_part, perm=[0,1,3,2])) / 2\r\n\r\n # Create a symmetric matrix for the imaginary part and positive definite\r\n imaginary_part = tf.random.uniform(shape, minval=0, maxval=1)\r\n imaginary_symmetric = (imaginary_part + tf.transpose(imaginary_part, perm=[0,1,3,2])) / 2\r\n\r\n # Combine into a complex tensor\r\n complex_tensor = tf.complex(real_symmetric, imaginary_symmetric)\r\n\r\n return complex_tensor\r\n\r\nparams = [\r\n generate_symmetric_positive_complex_tensor([1, 26, 2, 2]),\r\n # tf.complex(tf.random.uniform([1, 26, 2, 2], minval=0, maxval=1, dtype=tf.float32), tf.random.uniform([1, 26, 2, 2],minval=0, maxval=1, dtype=tf.float32)),\r\n]\r\n\r\nclass Model1(tf.keras.Model):\r\n def __init__(self):\r\n super().__init__()\r\n self.p0 = tf.constant(params[0]) # [1, 26, 2, 2] complex64\r\n\r\n @tf.function\r\n def __call__(self):\r\n\r\n cho = tf.linalg.cholesky(self.p0)\r\n return cho\r\n \r\nmodel1 = Model1()\r\n\r\ntf.config.run_functions_eagerly(True)\r\nout1 = model1()\r\nout2 = model1()\r\nprint(f'=========eager_output(version:{tf.__version__})================')\r\ntry:\r\n for i in range(len(out1)):\r\n np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.01, err_msg=f'at checking {i}th')\r\n print(\"Eager does not trigger assertion\")\r\nexcept AssertionError as e:\r\n print(\"Eager triggers assertion\")\r\n print(e)\r\n```", "@SuryanarayanaY I was able to replicate this issue on colab, please find the gist [here](https://colab.research.google.com/gist/sushreebarsa/4d9cfa90d00c6437c787cc80d43abe11/62451.ipynb#scrollTo=P1R23a332b52) for reference.\r\nThank you!" ]	2023-11-21T15:08:06	2023-11-28T04:36:14	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.15.0-dev20231005 ### Custom code Yes ### OS platform and distribution Ubuntu ### Mobile device _No response_ ### Python version 3.9 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I'm encountering a situation where a `tf.linalig.cholesky` produces different outputs for the same input in eager execution mode, which should theoretically yield consistent results. ### Standalone code to reproduce the issue ```python import tensorflow as tf import numpy as np params = [ tf.complex(tf.random.uniform([1, 26, 2, 2], dtype=tf.float32), tf.random.uniform([1, 26, 2, 2], dtype=tf.float32)), ] class Model1(tf.keras.Model): def __init__(self): super().__init__() self.p0 = tf.Variable(params[0]) # [1, 26, 2, 2] complex64 @tf.function def __call__(self, v7_0): cho = tf.linalg.cholesky(self.p0) return cho inputs = [ tf.complex(tf.random.uniform([1, 26, 1, 2], dtype=tf.float32), tf.random.uniform([1, 26, 1, 2], dtype=tf.float32)), ] model1 = Model1() with tf.device('cpu'): tf.config.run_functions_eagerly(True) out1 = model1(inputs) out2 = model1(inputs) print(f'=========eager_output(version:{tf.__version__})================') try : for i in range(len(out1)): np.testing.assert_allclose(out1[i].numpy(), out2[i].numpy(), rtol=0.01, err_msg=f'at checking {i}th') print("Eager does not trigger assertion") except AssertionError as e: print("Eeager triggers assertion") print(e) ``` ### Relevant log output ```shell 2023-11-21 23:04:32.635629: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered 2023-11-21 23:04:32.635665: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered 2023-11-21 23:04:32.637083: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered 2023-11-21 23:04:32.645338: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. 2023-11-21 23:04:33.736439: W tensorflow/compiler/tf2tensorrt/utils/py_utils.cc:38] TF-TRT Warning: Could not find TensorRT 2023-11-21 23:04:35.768320: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1924] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 4170 MB memory: -> device: 0, name: NVIDIA GeForce RTX 2070, pci bus id: 0000:02:00.0, compute capability: 7.5 2023-11-21 23:04:35.769152: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1924] Created device /job:localhost/replica:0/task:0/device:GPU:1 with 6678 MB memory: -> device: 1, name: NVIDIA GeForce RTX 2070, pci bus id: 0000:04:00.0, compute capability: 7.5 2023-11-21 23:04:35.769846: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1924] Created device /job:localhost/replica:0/task:0/device:GPU:2 with 6674 MB memory: -> device: 2, name: NVIDIA GeForce RTX 2070, pci bus id: 0000:83:00.0, compute capability: 7.5 2023-11-21 23:04:35.770891: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1924] Created device /job:localhost/replica:0/task:0/device:GPU:3 with 6260 MB memory: -> device: 3, name: NVIDIA GeForce RTX 2070, pci bus id: 0000:84:00.0, compute capability: 7.5 2023-11-21 23:04:35.928481: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928511: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928518: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928524: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928530: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928541: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928553: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928561: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928567: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928573: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928578: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928584: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928590: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928595: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928601: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928606: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928613: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928623: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928633: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928640: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928645: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928651: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928656: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928881: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928896: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928902: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928907: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928913: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928918: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928923: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928929: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928934: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928940: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928945: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928951: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928956: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928962: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928967: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928973: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928978: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928983: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928988: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.928994: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.929000: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.929011: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. 2023-11-21 23:04:35.929022: W tensorflow/core/kernels/linalg/cholesky_op.cc:56] Cholesky decomposition was not successful. Eigen::LLT failed with error code 1. Filling lower-triangular output with NaNs. =========eager_output(version:2.15.0-dev20231005)================ Eeager triggers assertion Not equal to tolerance rtol=0.01, atol=0 at checking 0th x and y nan location mismatch: x: array([[[0. +0.j , 0. +0.j ], [0. +0.j , 0. +0.j ]], ... y: array([[[0.000000e+00+0.j , 1.401298e-45 +nanj], [0.000000e+00+0.j , 0.000000e+00+0.j ]], ... ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62451/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62451/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62450	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62450/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62450/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62450/events	https://github.com/tensorflow/tensorflow/issues/62450	2,004,368,289	I_kwDOArmXAs53eDuh	62,450	Custom train_step in subclass model not saved	{ "login": "rcalonso", "id": 19788974, "node_id": "MDQ6VXNlcjE5Nzg4OTc0", "avatar_url": "https://avatars.githubusercontent.com/u/19788974?v=4", "gravatar_id": "", "url": "https://api.github.com/users/rcalonso", "html_url": "https://github.com/rcalonso", "followers_url": "https://api.github.com/users/rcalonso/followers", "following_url": "https://api.github.com/users/rcalonso/following{/other_user}", "gists_url": "https://api.github.com/users/rcalonso/gists{/gist_id}", "starred_url": "https://api.github.com/users/rcalonso/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/rcalonso/subscriptions", "organizations_url": "https://api.github.com/users/rcalonso/orgs", "repos_url": "https://api.github.com/users/rcalonso/repos", "events_url": "https://api.github.com/users/rcalonso/events{/privacy}", "received_events_url": "https://api.github.com/users/rcalonso/received_events", "type": "User", "site_admin": false }	[ { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 473184161, "node_id": "MDU6TGFiZWw0NzMxODQxNjE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:support", "name": "type:support", "color": "159b2e", "default": false, "description": "Support issues" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @rcalonso ,\r\n\r\nKeras has introduced a decorator `@keras.saving.register_keras_serializable() ` which needs to be decorated on your custom class and save format should be `.keras`.\r\n\r\nI have done the required changes and tested with Tf2.14 and keras-nightly(3.0.0.Dev) and both working fine and able to reload custom train_step properly.Please refer attached gists [Tf2.14](https://colab.sandbox.google.com/gist/SuryanarayanaY/38b7e326ca9139f59dcc81b4d055788b/62450-tf2-15.ipynb) and [keras3](https://colab.sandbox.google.com/gist/SuryanarayanaY/b78b698e83a39069e1a536c29836b6c8/62450-keras3-0-dev.ipynb).\r\n\r\nHope this resolves your query.\r\n\r\n", "Hi @SuryanarayanaY,\r\n\r\nMany thanks! I tried before with the decorator `@keras.saving.register_keras_serializable()` but I was saving in `tf` format. Using the save format `.keras` works as expected.\r\n\r\nThanks for the help.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62450\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62450\">No</a>\n" ]	2023-11-21T14:07:41	2023-11-22T15:10:02	2023-11-22T15:10:00	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source binary ### TensorFlow version 2.13.1 \| 2.14.0 \| 2.15.0 ### Custom code Yes ### OS platform and distribution Google Colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Hi, Describe the current behavior I'm building a custom model via model subclassing following the steps in https://www.tensorflow.org/guide/keras/customizing_what_happens_in_fit. I'm overriding the `train_step` and `test_step` method. I can train the model without any problem but when I load a saved model, the `train_step` and `test_step` are not loaded and instead the methods from the super class `tf.keras.Model` are used. I've tried the suggestions from previous issues where they had the same problem but the suggestions don't work. - https://github.com/tensorflow/tensorflow/issues/38103 - https://github.com/tensorflow/tensorflow/issues/46530 Also, I've gone through the documentation about serialization and saving https://www.tensorflow.org/guide/keras/serialization_and_saving#how_savedmodel_handles_custom_objects and the suggestion from https://github.com/tensorflow/tensorflow/issues/46530#issuecomment-790209731 but it doesn't work. Describe the expected behavior I'd expect that `train_step` and `test_step` are loaded from the saved model. ### Standalone code to reproduce the issue https://gist.github.com/rcalonso/8b0c55f3e31b294a9fe68d0f1f277635 ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62450/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62450/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62449	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62449/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62449/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62449/events	https://github.com/tensorflow/tensorflow/issues/62449	2,004,283,080	I_kwDOArmXAs53du7I	62,449	Encountering an error when using " tf.config.experimental.enable_op_determinism()	{ "login": "k1379432", "id": 85670902, "node_id": "MDQ6VXNlcjg1NjcwOTAy", "avatar_url": "https://avatars.githubusercontent.com/u/85670902?v=4", "gravatar_id": "", "url": "https://api.github.com/users/k1379432", "html_url": "https://github.com/k1379432", "followers_url": "https://api.github.com/users/k1379432/followers", "following_url": "https://api.github.com/users/k1379432/following{/other_user}", "gists_url": "https://api.github.com/users/k1379432/gists{/gist_id}", "starred_url": "https://api.github.com/users/k1379432/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/k1379432/subscriptions", "organizations_url": "https://api.github.com/users/k1379432/orgs", "repos_url": "https://api.github.com/users/k1379432/repos", "events_url": "https://api.github.com/users/k1379432/events{/privacy}", "received_events_url": "https://api.github.com/users/k1379432/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1097545817, "node_id": "MDU6TGFiZWwxMDk3NTQ1ODE3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:apis", "name": "comp:apis", "color": "0052cc", "default": false, "description": "Highlevel API related issues" }, { "id": 4511033337, "node_id": "LA_kwDOArmXAs8AAAABDODn-Q", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.10", "name": "TF 2.10", "color": "C15088", "default": false, "description": "" } ]	closed	false	{ "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false }	[ { "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false } ]	null	[ "@k1379432 Please try to upgrade to the latest TF version. I tried it on colab and didn't face any issue, please find the gist [here](https://colab.research.google.com/gist/sushreebarsa/20f9ca2a12c62aec2f71db21b244a951/62449.ipynb). We need to check whether it is an issue with windows OS ? \r\nThank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62449\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62449\">No</a>\n" ]	2023-11-21T13:24:05	2023-12-12T01:49:53	2023-12-12T01:49:49	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf2.10 ### Custom code Yes ### OS platform and distribution Windows ### Mobile device Windows ### Python version 3.7 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version cuDNN11.2/CUDA 11.2 ### GPU model and memory NVIDIA GeForce RTX 2080 Ti ### Current behavior? "I'm encountering an error in the optimizer_v2.py file when using " tf.config.experimental.enable_op_determinism() ", and I'm unsure how to resolve it ### Standalone code to reproduce the issue ```shell from tensorflow.keras.models import * from tensorflow.keras.layers import * from tensorflow.keras.applications import ResNet50, ResNet101, ResNet152, NASNetMobile #from tensorflow.keras.applications import EfficientNetB4, EfficientNetB5, EfficientNetV2S, ConvNeXtBase import tensorflow.keras.backend as K import tensorflow as tf import random import os def random_seed(seed): os.environ['PYTHONHASHSEED'] = str(seed) # Python general random.seed(seed) # Python random tf.random.set_seed(seed) tf.keras.utils.set_random_seed(seed) tf.compat.v1.set_random_seed(42) os.environ['TF_CUDNN_DETERMINISTIC'] = '1' tf.config.experimental.enable_op_determinism() random_seed(42) ``` ### Relevant log output ```shell 2023-11-21 20:44:13.447175: I tensorflow/stream_executor/cuda/cuda_dnn.cc:384] Loaded cuDNN version 8100 Traceback (most recent call last): File "training2.py", line 155, in <module> callbacks=[ checkpoint, logger, early_stop ]) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\utils\traceback_utils.py", line 67, in error_handler raise e.with_traceback(filtered_tb) from None File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\tensorflow\python\eager\execute.py", line 55, in quick_execute inputs, attrs, num_outputs) tensorflow.python.framework.errors_impl.UnimplementedError: Graph execution error: Detected at node 'gradient_tape/model/FPN2.upsample/resize/ResizeNearestNeighborGrad' defined at (most recent call last): File "training2.py", line 155, in <module> callbacks=[ checkpoint, logger, early_stop ]) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\utils\traceback_utils.py", line 64, in error_handler return fn(args, *kwargs) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\engine\training.py", line 1409, in fit tmp_logs = self.train_function(iterator) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\engine\training.py", line 1051, in train_function return step_function(self, iterator) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\engine\training.py", line 1040, in step_function outputs = model.distribute_strategy.run(run_step, args=(data,)) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\engine\training.py", line 1030, in run_step outputs = model.train_step(data) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\engine\training.py", line 893, in train_step self.optimizer.minimize(loss, self.trainable_variables, tape=tape) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\optimizers\optimizer_v2\optimizer_v2.py", line 538, in minimize loss, var_list=var_list, grad_loss=grad_loss, tape=tape) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\optimizers\optimizer_v2\optimizer_v2.py", line 590, in _compute_gradients grads_and_vars = self._get_gradients(tape, loss, var_list, grad_loss) File "C:\Users\LAB\anaconda3\envs\test2\lib\site-packages\keras\optimizers\optimizer_v2\optimizer_v2.py", line 471, in _get_gradients grads = tape.gradient(loss, var_list, grad_loss) Node: 'gradient_tape/model/FPN2.upsample/resize/ResizeNearestNeighborGrad' A deterministic GPU implementation of ResizeNearestNeighborGrad is not currently available. [[{{node gradient_tape/model/FPN2.upsample/resize/ResizeNearestNeighborGrad}}]] [Op:__inference_train_function_36513] 2023-11-21 20:44:16.419233: W tensorflow/core/kernels/data/generator_dataset_op.cc:108] Error occurred when finalizing GeneratorDataset iterator: FAILED_PRECONDITION: Python interpreter state is not initialized. The process may be terminated. [[{{node PyFunc}}]] ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62449/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62449/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62448	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62448/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62448/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62448/events	https://github.com/tensorflow/tensorflow/issues/62448	2,004,259,333	I_kwDOArmXAs53dpIF	62,448	manylinux_2_28 support	{ "login": "ratan-prasad", "id": 82648972, "node_id": "MDQ6VXNlcjgyNjQ4OTcy", "avatar_url": "https://avatars.githubusercontent.com/u/82648972?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ratan-prasad", "html_url": "https://github.com/ratan-prasad", "followers_url": "https://api.github.com/users/ratan-prasad/followers", "following_url": "https://api.github.com/users/ratan-prasad/following{/other_user}", "gists_url": "https://api.github.com/users/ratan-prasad/gists{/gist_id}", "starred_url": "https://api.github.com/users/ratan-prasad/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ratan-prasad/subscriptions", "organizations_url": "https://api.github.com/users/ratan-prasad/orgs", "repos_url": "https://api.github.com/users/ratan-prasad/repos", "events_url": "https://api.github.com/users/ratan-prasad/events{/privacy}", "received_events_url": "https://api.github.com/users/ratan-prasad/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 1205615612, "node_id": "MDU6TGFiZWwxMjA1NjE1NjEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:%20ubuntu/linux", "name": "subtype: ubuntu/linux", "color": "b619ea", "default": false, "description": "Ubuntu/Linux Build/Installation Issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	open	false	{ "login": "angerson", "id": 32465472, "node_id": "MDQ6VXNlcjMyNDY1NDcy", "avatar_url": "https://avatars.githubusercontent.com/u/32465472?v=4", "gravatar_id": "", "url": "https://api.github.com/users/angerson", "html_url": "https://github.com/angerson", "followers_url": "https://api.github.com/users/angerson/followers", "following_url": "https://api.github.com/users/angerson/following{/other_user}", "gists_url": "https://api.github.com/users/angerson/gists{/gist_id}", "starred_url": "https://api.github.com/users/angerson/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/angerson/subscriptions", "organizations_url": "https://api.github.com/users/angerson/orgs", "repos_url": "https://api.github.com/users/angerson/repos", "events_url": "https://api.github.com/users/angerson/events{/privacy}", "received_events_url": "https://api.github.com/users/angerson/received_events", "type": "User", "site_admin": false }	[ { "login": "angerson", "id": 32465472, "node_id": "MDQ6VXNlcjMyNDY1NDcy", "avatar_url": "https://avatars.githubusercontent.com/u/32465472?v=4", "gravatar_id": "", "url": "https://api.github.com/users/angerson", "html_url": "https://github.com/angerson", "followers_url": "https://api.github.com/users/angerson/followers", "following_url": "https://api.github.com/users/angerson/following{/other_user}", "gists_url": "https://api.github.com/users/angerson/gists{/gist_id}", "starred_url": "https://api.github.com/users/angerson/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/angerson/subscriptions", "organizations_url": "https://api.github.com/users/angerson/orgs", "repos_url": "https://api.github.com/users/angerson/repos", "events_url": "https://api.github.com/users/angerson/events{/privacy}", "received_events_url": "https://api.github.com/users/angerson/received_events", "type": "User", "site_admin": false }, { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "This was discussed at a SIG Build meeting recently (http://bit.ly/tf-sig-build-notes). Will likely be discussed again on the next one in December", "Thanks @mihaimaruseac for sharing the SIG build meeting pointers. We will wait when this is decided. For short term, I have one specific questions. If we install GCC11 in manylinux2014, will it break/violate manylinux2014 rules? Do you foresee any issue in such upgrade of gcc version?", "My understanding is that unless you compile everything with the same compiler you might hit some hard to debug issues (data corruption, etc.). Maybe not if you're using newer compilers now, but it was a certain thing in the past if you had some piece of code that used one ABI for `std::string` and another piece that used another ABI.\r\n\r\nIf you build everything yourself, you won't have this issue, but it will no longer be manylinux2014.", "Thanks @mihaimaruseac for the reply.\r\nI have created manylinux2014 env and installed gcc 11 using the command:\r\nyum install -y centos-release-scl\r\nyum install -y devtoolset-11\r\nNow if i am building TensorFlow for CPU using the default bazel build command, do you think it will be no longer manylinux2014?\r\nIs there any workaround for it?\r\n\r\n", "I'm not sure. You can try building the wheel and then run `auditwheel` on it to see manylinux compliance" ]	2023-11-21T13:12:52	2023-12-11T16:21:23	null	NONE	null	null	null	### Issue type Build/Install ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version TF 2.15 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04 ### Mobile device Linux Ubuntu 22.04 ### Python version 3.9 ### Bazel version Latest ### GCC/compiler version 10.2 ### CUDA/cuDNN version NA ### GPU model and memory NA ### Current behavior? This is not a bug, but didnt find a better place to ask the question. Can we use manylinux_2_28 to build latest tensorflow v2.15? Latest binaries are released with manylinux2014 which use gcc 10.2.x. However to use the some of the flags available in GCC11/GCC12, can we use manylinux_2_28 to build Tensorflow binaries for CPU? ### Standalone code to reproduce the issue ```shell N/A ``` ### Relevant log output ```shell NA ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62448/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62448/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62447	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62447/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62447/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62447/events	https://github.com/tensorflow/tensorflow/issues/62447	2,004,126,731	I_kwDOArmXAs53dIwL	62,447	StridedSlice tests have invalid output data	{ "login": "davmon01", "id": 44364347, "node_id": "MDQ6VXNlcjQ0MzY0MzQ3", "avatar_url": "https://avatars.githubusercontent.com/u/44364347?v=4", "gravatar_id": "", "url": "https://api.github.com/users/davmon01", "html_url": "https://github.com/davmon01", "followers_url": "https://api.github.com/users/davmon01/followers", "following_url": "https://api.github.com/users/davmon01/following{/other_user}", "gists_url": "https://api.github.com/users/davmon01/gists{/gist_id}", "starred_url": "https://api.github.com/users/davmon01/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/davmon01/subscriptions", "organizations_url": "https://api.github.com/users/davmon01/orgs", "repos_url": "https://api.github.com/users/davmon01/repos", "events_url": "https://api.github.com/users/davmon01/events{/privacy}", "received_events_url": "https://api.github.com/users/davmon01/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false }	[ { "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false } ]	null	[ "@davmon01 As a workaround, could you please try to use different slicing operations such as [tf.slice](https://www.tensorflow.org/api_docs/python/tf/slice) or [tf.gather](https://www.tensorflow.org/api_docs/python/tf/gather) ?\r\n@LakshmiKalaKadali Could you please have a look at this ticket?\r\nThank you!", "Hi @davmon01, can you give me the commands you used to do this?:\r\n\r\n```\r\nThis was reproduced by compiling the Delegate Test Suite with the ArmNN Delegate for Tensorflow Lite and attempting to run those tests against the ArmNN delegate.\r\n```\r\n\r\nThanks for your help.", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62447\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62447\">No</a>\n" ]	2023-11-21T12:02:59	2023-12-20T08:34:11	2023-12-15T01:49:20	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.14 ### Custom code Yes ### OS platform and distribution Ubuntu 20.04 ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Strided Slice tests under: tensorflow/lite/kernels/strided_slice_test.cc have invalid output data. They appear to have a slice size that is 1 larger than it should be given the definition in the documentation. From: https://www.tensorflow.org/api_docs/python/tf/strided_slice A slice is given as size (end-begin)/stride from the given input_ tensor. But as an example in the following test: TYPED_TEST(StridedSliceOpTest, Offset) { for (bool constant_tensors : {true, false}) { if (SingleOpModel::GetForceUseNnapi() && constant_tensors) { // NNAPI does not support graphs with all constant inputs. continue; } StridedSliceOpModel<TypeParam> m( {10}, {1}, {1}, {1}, std::vector<TypeParam>{0, 1, 2, 3, 4, 5, 6, 7, 8, 9}, {1}, {3}, {1}, 0, 0, 0, 0, 0, constant_tensors, /offset=/true); ASSERT_EQ(m.Invoke(), kTfLiteOk); EXPECT_THAT(m.GetOutputShape(), ElementsAreArray({3})); EXPECT_THAT(m.GetOutput(), ElementsAreArray({1, 2, 3})); if (constant_tensors) { EXPECT_THAT(m.GetOutputTensor(0)->allocation_type, kTfLitePersistentRo); } else { EXPECT_THAT(m.GetOutputTensor(0)->allocation_type, kTfLiteArenaRw); } } } Where you would expect a slice of length 2, it instead expects 3 elements. This seems to be the case for all of the tests in the log below. ### Standalone code to reproduce the issue ```shell This was reproduced by compiling the Delegate Test Suite with the ArmNN Delegate for Tensorflow Lite and attempting to run those tests against the ArmNN delegate. Based on the nature of the error though, we would like clarification on if the expected output data is in fact correct and the documentation regarding the slice size is wrong. If on the other hand the data is wrong and the documentation is correct, then any method of running these tests contained within tensorflow/lite/kernels/strided_slice_test.cc will give the incorrect results. ``` ### Relevant log output ```shell [ FAILED ] StridedSliceOpTest/0.In1DEmpty, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.Offset, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.OffsetArray, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.OffsetConstant, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.OffsetConstantStride, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.OffsetConstantNegativeStride, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.In5D_Identity, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.In5D_IdentityShrinkAxis1, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.In3D_Backward, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.EllipsisMask1_NewAxisMask2, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.EllipsisMask2_NewAxisMask1, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.EllipsisMask2_NewAxisMask5, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.EllipsisMask2_NewAxisMask2, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.EllipsisMask4_NewAxisMask2, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.EllipsisMask2, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.NewAxisMask2, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.NewAxisMask1, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.OneOneOne, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.StrideOutOfBounds, where TypeParam = float [ FAILED ] StridedSliceOpTest/0.NoopOffset, where TypeParam = float [ FAILED ] StridedSliceOpTest/1.In1DEmpty, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.Offset, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.OffsetArray, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.OffsetConstant, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.OffsetConstantStride, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.OffsetConstantNegativeStride, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.In5D_Identity, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.In5D_IdentityShrinkAxis1, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.In3D_Backward, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.EllipsisMask1_NewAxisMask2, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.EllipsisMask2_NewAxisMask1, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.EllipsisMask2_NewAxisMask5, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.EllipsisMask2_NewAxisMask2, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.EllipsisMask4_NewAxisMask2, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.EllipsisMask2, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.NewAxisMask2, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.NewAxisMask1, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.OneOneOne, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.StrideOutOfBounds, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/1.NoopOffset, where TypeParam = unsigned char [ FAILED ] StridedSliceOpTest/3.In1DEmpty, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.Offset, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.OffsetArray, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.OffsetConstant, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.OffsetConstantStride, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.OffsetConstantNegativeStride, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.In5D_Identity, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.In5D_IdentityShrinkAxis1, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.In3D_Backward, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.EllipsisMask1_NewAxisMask2, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.EllipsisMask2_NewAxisMask1, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.EllipsisMask2_NewAxisMask5, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.EllipsisMask2_NewAxisMask2, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.EllipsisMask4_NewAxisMask2, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.EllipsisMask2, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.NewAxisMask2, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.NewAxisMask1, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.OneOneOne, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.StrideOutOfBounds, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/3.NoopOffset, where TypeParam = signed char [ FAILED ] StridedSliceOpTest/4.In1DEmpty, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.Offset, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.OffsetArray, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.OffsetConstant, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.OffsetConstantStride, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.OffsetConstantNegativeStride, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.In5D_Identity, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.In5D_IdentityShrinkAxis1, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.In3D_Backward, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.EllipsisMask1_NewAxisMask2, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.EllipsisMask2_NewAxisMask1, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.EllipsisMask2_NewAxisMask5, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.EllipsisMask2_NewAxisMask2, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.EllipsisMask4_NewAxisMask2, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.EllipsisMask2, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.NewAxisMask2, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.NewAxisMask1, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.OneOneOne, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.StrideOutOfBounds, where TypeParam = short [ FAILED ] StridedSliceOpTest/4.NoopOffset, where TypeParam = short ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62447/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62447/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62446	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62446/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62446/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62446/events	https://github.com/tensorflow/tensorflow/pull/62446	2,004,126,441	PR_kwDOArmXAs5gAzhC	62,446	Fix misuse of ArrayRef	{ "login": "elfringham", "id": 10442001, "node_id": "MDQ6VXNlcjEwNDQyMDAx", "avatar_url": "https://avatars.githubusercontent.com/u/10442001?v=4", "gravatar_id": "", "url": "https://api.github.com/users/elfringham", "html_url": "https://github.com/elfringham", "followers_url": "https://api.github.com/users/elfringham/followers", "following_url": "https://api.github.com/users/elfringham/following{/other_user}", "gists_url": "https://api.github.com/users/elfringham/gists{/gist_id}", "starred_url": "https://api.github.com/users/elfringham/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/elfringham/subscriptions", "organizations_url": "https://api.github.com/users/elfringham/orgs", "repos_url": "https://api.github.com/users/elfringham/repos", "events_url": "https://api.github.com/users/elfringham/events{/privacy}", "received_events_url": "https://api.github.com/users/elfringham/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169365494, "node_id": "MDU6TGFiZWwxMTY5MzY1NDk0", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:M", "name": "size:M", "color": "adafea", "default": false, "description": "CL Change Size: Medium" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "@gbaned Thanks for triggering the ARM_CI run on this PR. Unfortunately it failed due to an unrelated API change issue. I have just rebased the code to pick up the fix for the API change.", "@GleasonK Please could you review this again now?" ]	2023-11-21T12:02:49	2023-12-12T14:29:34	2023-12-12T03:10:52	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62446", "html_url": "https://github.com/tensorflow/tensorflow/pull/62446", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62446.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62446.patch", "merged_at": "2023-12-12T03:10:52" }	Fix some uses of ArrayRef that result in out of scope accesses being made. This includes accesses on the stack after being freed. Also fixed a concat attempt of two arrays through ArrayRef that was producing nonsense. The issues were located with the address sanitizer and shown to be fixed with these changes.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62446/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62446/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62445	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62445/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62445/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62445/events	https://github.com/tensorflow/tensorflow/issues/62445	2,003,660,837	I_kwDOArmXAs53bXAl	62,445	Using bazel build a c api	{ "login": "panhu", "id": 11703018, "node_id": "MDQ6VXNlcjExNzAzMDE4", "avatar_url": "https://avatars.githubusercontent.com/u/11703018?v=4", "gravatar_id": "", "url": "https://api.github.com/users/panhu", "html_url": "https://github.com/panhu", "followers_url": "https://api.github.com/users/panhu/followers", "following_url": "https://api.github.com/users/panhu/following{/other_user}", "gists_url": "https://api.github.com/users/panhu/gists{/gist_id}", "starred_url": "https://api.github.com/users/panhu/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/panhu/subscriptions", "organizations_url": "https://api.github.com/users/panhu/orgs", "repos_url": "https://api.github.com/users/panhu/repos", "events_url": "https://api.github.com/users/panhu/events{/privacy}", "received_events_url": "https://api.github.com/users/panhu/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 473184161, "node_id": "MDU6TGFiZWw0NzMxODQxNjE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:support", "name": "type:support", "color": "159b2e", "default": false, "description": "Support issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @panhu ,\r\n\r\nOn linux given build command should generate `libtensorflowlite_c.so`, and the mentioned size seems Ok to me as per the similar ticket #48068. \r\n\r\nYou may also test with C-Make build as per instructions [here](https://www.tensorflow.org/lite/guide/build_cmake).\r\n\r\nYou can also refer lite guide [here](https://www.tensorflow.org/lite/guide#:~:text=Key%20Point%3A%20The%20TensorFlow%20Lite%20binary%20is%20~1MB%20when%20all%20125%2B%20supported%20operators%20are%20linked%20(for%2032%2Dbit%20ARM%20builds)%2C%20and%20less%20than%20300KB%20when%20using%20only%20the%20operators%20needed%20for%20supporting%20the%20common%20image%20classification%20models%20InceptionV3%20and%20MobileNet.). Basically all tensorflow Ops are not supported for tflite and hence it is lightweight.\r\n\r\n", "Hi@SuryanarayanaY:\r\nThank you for your reply. I would like to ask if Tflite supports the Risc-v architecture. If so, how should we build the relevant shared library of tflite ? Is there a Risc-v toolchain?\r\nThanks!", "Hi @panhu,\r\nIt seems this query is already answered [here](https://github.com/tensorflow/tensorflow/issues/60849#issuecomment-1633217814).", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62445\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62445\">No</a>\n" ]	2023-11-21T07:51:04	2023-12-09T01:48:14	2023-12-09T01:48:10	NONE	null	null	null	System information - OS Platform and Distribution Linux Centos8): - TensorFlow installed from source - TensorFlow version ( github SHA if from source): Question Hi: When I used 'bazel build -c opt //tensorflow/lite/c:tensorflowlite_c' to create the C API for tflite. I only got 'libtensorflowlite_c. so' of only 4.4M, and I'm not sure if it's correct because it's too small?May I ask if this size is correct? Why is it so small? Thanks!	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62445/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62445/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62444	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62444/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62444/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62444/events	https://github.com/tensorflow/tensorflow/issues/62444	2,003,620,687	I_kwDOArmXAs53bNNP	62,444	tensorflow - tf.keras.Model.fit causes run out of data for validation data with validation_steps being set	{ "login": "oonisim", "id": 15814603, "node_id": "MDQ6VXNlcjE1ODE0NjAz", "avatar_url": "https://avatars.githubusercontent.com/u/15814603?v=4", "gravatar_id": "", "url": "https://api.github.com/users/oonisim", "html_url": "https://github.com/oonisim", "followers_url": "https://api.github.com/users/oonisim/followers", "following_url": "https://api.github.com/users/oonisim/following{/other_user}", "gists_url": "https://api.github.com/users/oonisim/gists{/gist_id}", "starred_url": "https://api.github.com/users/oonisim/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/oonisim/subscriptions", "organizations_url": "https://api.github.com/users/oonisim/orgs", "repos_url": "https://api.github.com/users/oonisim/repos", "events_url": "https://api.github.com/users/oonisim/events{/privacy}", "received_events_url": "https://api.github.com/users/oonisim/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@sachinprasadhs I was able to replicate the issue on colab, please find the [gist](https://colab.research.google.com/gist/sushreebarsa/654d0e421b5dd863569ecdd897870e4c/62444.ipynb) here for reference.\r\nThank you!", "Hi, \r\n\r\nPlease go through the discussion [here](https://stackoverflow.com/questions/59864408/tensorflowyour-input-ran-out-of-data) to understand the correct number of steps for validation and training.\r\nBasic rule to follow is:\r\n```\r\nsteps_per_epoch = len(X_train)//batch_size\r\n\r\nvalidation_steps = len(X_test)//batch_size\r\n```\r\n", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62444\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62444\">No</a>\n" ]	2023-11-21T07:24:17	2023-12-22T01:48:34	2023-12-22T01:48:30	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source binary ### TensorFlow version 2.14.1 ### Custom code No ### OS platform and distribution Ubuntu 22.04 LTS ### Mobile device _No response_ ### Python version 3.10.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version 11.8 ### GPU model and memory _No response_ ### Current behavior? Trying to understand the ```validation_steps``` parameter of [tf.keras.Model.fit](https://www.tensorflow.org/api_docs/python/tf/keras/Model#fit). > Total number of steps (batches of samples) to draw before stopping when performing validation at the end of every epoch. For instance, [TFDS MNIST](https://www.tensorflow.org/datasets/catalog/mnist) dataset has ```60,000``` train and ```10,000``` test data records. Trying to consume all the records during ```num_epochs=2``` epochs with ```batch_size=8``` using generators as the data sources to the model. ``` (train, test), info = tfds.load( 'mnist', split=['train', 'test'], shuffle_files=True, as_supervised=True, with_info=True, ) x_generator = train.batch(batch_size).as_numpy_iterator() v_generator = test.batch(batch_size).as_numpy_iterator() # using 'test' for validation here ``` The training data can afford ```3750=(60000 / batch_size=8 / epochs=2)``` batches, and the test data can afford ```625=(10000 / batch_size=8 / epochs=2)``` batches. ``` def f(image, label): return 1 num_total_train_records = len(list( # 60000 train.map(f) )) num_total_test_records = len(list( # 10000 test.map(f) )) print(num_total_train_records, num_total_test_records) ----- 60000 10000 ``` ``` num_epochs = 2 batch_size = 8 num_x_batches_per_epoch = int(np.floor(num_total_train_records / batch_size / num_epochs)) num_v_batches_per_epoch = int(np.floor(num_total_test_records / batch_size / num_epochs)) print(num_x_batches_per_epoch, num_v_batches_per_epoch) # --- # show 3750 625 ``` However, setting ```tf.keras.Model.fit(validation_steps=625)``` causes the error ```Your input ran out of data... Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 625 batches)```. ``` model.fit( x=x_generator , epochs=num_epochs, batch_size=batch_size, steps_per_epoch=num_x_batches_per_epoch, validation_data=v_generator, validation_steps=num_v_batches_per_epoch, validation_batch_size=batch_size ) ``` ``` Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 625 batches). You may need to use the repeat() function when building your dataset. 2023-11-21 17:39:33.226528: I tensorflow/core/framework/local_rendezvous.cc:421] Local rendezvous recv item cancelled. Key hash: 17391114698345974101 2023-11-21 17:39:33.226580: I tensorflow/core/framework/local_rendezvous.cc:421] Local rendezvous recv item cancelled. Key hash: 8226056677969075330 WARNING:tensorflow:Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 625 batches). You may need to use the repeat() function when building your dataset. ``` ## Code ``` import numpy as np import tensorflow as tf from tensorflow import keras import tensorflow_datasets as tfds (train, test), info = tfds.load( 'mnist', split=['train', 'test'], shuffle_files=True, as_supervised=True, with_info=True, ) def f(image, label): return 1 num_total_train_records = len(list( train.map(f) )) num_total_test_records = len(list( test.map(f) )) print(num_total_train_records, num_total_test_records) model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(input_shape=(28, 28)), tf.keras.layers.Dense(128, activation='relu'), tf.keras.layers.Dense(10) ]) model.compile( optimizer=tf.keras.optimizers.Adam(0.001), loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True), metrics=[tf.keras.metrics.SparseCategoricalAccuracy()], ) num_epochs = 2 batch_size = 8 num_x_batches_per_epoch = int(np.floor(num_total_train_records / batch_size / num_epochs)) num_v_batches_per_epoch = int(np.floor(num_total_test_records / batch_size / num_epochs)) print(num_x_batches_per_epoch, num_v_batches_per_epoch) # --- # will show 3750 625 x_generator = train.batch(batch_size).as_numpy_iterator() v_generator = test.batch(batch_size).as_numpy_iterator() model.fit( x=x_generator , epochs=num_epochs, batch_size=batch_size, steps_per_epoch=num_x_batches_per_epoch, validation_data=v_generator, validation_steps=num_v_batches_per_epoch, validation_batch_size=batch_size ) ``` By minus 1, it works. ``` num_v_batches_per_epoch = int(np.floor(num_total_test_records / batch_size / num_epochs)) -1 # Cuase ran out of data without -1 ``` Please help understand this behavior. Also the document says ```Only relevant if validation_data is provided and is a tf.data dataset.``` but obviously it is not only for ```tf.data.Dataset```. ## Environment ``` tensorflow 2.14.1 Python 3.10.12 Ubuntu 22.04 LTS ``` [1]: https://www.tensorflow.org/datasets/catalog/mnist ### Standalone code to reproduce the issue ```shell import numpy as np import tensorflow as tf from tensorflow import keras import tensorflow_datasets as tfds (train, test), info = tfds.load( 'mnist', split=['train', 'test'], shuffle_files=True, as_supervised=True, with_info=True, ) def f(image, label): return 1 num_total_train_records = len(list( train.map(f) )) num_total_test_records = len(list( test.map(f) )) print(num_total_train_records, num_total_test_records) model = tf.keras.models.Sequential([ tf.keras.layers.Flatten(input_shape=(28, 28)), tf.keras.layers.Dense(128, activation='relu'), tf.keras.layers.Dense(10) ]) model.compile( optimizer=tf.keras.optimizers.Adam(0.001), loss=tf.keras.losses.SparseCategoricalCrossentropy(from_logits=True), metrics=[tf.keras.metrics.SparseCategoricalAccuracy()], ) num_epochs = 2 batch_size = 8 num_x_batches_per_epoch = int(np.floor(num_total_train_records / batch_size / num_epochs)) num_v_batches_per_epoch = int(np.floor(num_total_test_records / batch_size / num_epochs)) print(num_x_batches_per_epoch, num_v_batches_per_epoch) # --- # will show 3750 625 x_generator = train.batch(batch_size).as_numpy_iterator() v_generator = test.batch(batch_size).as_numpy_iterator() model.fit( x=x_generator , epochs=num_epochs, batch_size=batch_size, steps_per_epoch=num_x_batches_per_epoch, validation_data=v_generator, validation_steps=num_v_batches_per_epoch, validation_batch_size=batch_size ) ``` ### Relevant log output ```shell Epoch 1/2 3750/3750 [==============================] - 13s 3ms/step - loss: 0.3059 - sparse_categorical_accuracy: 0.9338 - val_loss: 0.7984 - val_sparse_categorical_accuracy: 0.9214 Epoch 2/2 3748/3750 [============================>.] - ETA: 0s - loss: 0.3194 - sparse_categorical_accuracy: 0.9305WARNING:tensorflow:Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 625 batches). You may need to use the repeat() function when building your dataset. 2023-11-21 17:39:33.226528: I tensorflow/core/framework/local_rendezvous.cc:421] Local rendezvous recv item cancelled. Key hash: 17391114698345974101 2023-11-21 17:39:33.226580: I tensorflow/core/framework/local_rendezvous.cc:421] Local rendezvous recv item cancelled. Key hash: 8226056677969075330 WARNING:tensorflow:Your input ran out of data; interrupting training. Make sure that your dataset or generator can generate at least `steps_per_epoch * epochs` batches (in this case, 625 batches). You may need to use the repeat() function when building your dataset. 3750/3750 [==============================] - 13s 3ms/step - loss: 0.3194 - sparse_categorical_accuracy: 0.9305 - val_loss: 0.6366 - val_sparse_categorical_accuracy: 0.9075 ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62444/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62444/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62443	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62443/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62443/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62443/events	https://github.com/tensorflow/tensorflow/issues/62443	2,002,891,835	I_kwDOArmXAs53YbQ7	62,443	Bus error when importing keras_nlp before tensorflow	{ "login": "yaddayaddayaddayadda", "id": 45805059, "node_id": "MDQ6VXNlcjQ1ODA1MDU5", "avatar_url": "https://avatars.githubusercontent.com/u/45805059?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yaddayaddayaddayadda", "html_url": "https://github.com/yaddayaddayaddayadda", "followers_url": "https://api.github.com/users/yaddayaddayaddayadda/followers", "following_url": "https://api.github.com/users/yaddayaddayaddayadda/following{/other_user}", "gists_url": "https://api.github.com/users/yaddayaddayaddayadda/gists{/gist_id}", "starred_url": "https://api.github.com/users/yaddayaddayaddayadda/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yaddayaddayaddayadda/subscriptions", "organizations_url": "https://api.github.com/users/yaddayaddayaddayadda/orgs", "repos_url": "https://api.github.com/users/yaddayaddayaddayadda/repos", "events_url": "https://api.github.com/users/yaddayaddayaddayadda/events{/privacy}", "received_events_url": "https://api.github.com/users/yaddayaddayaddayadda/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 1205765054, "node_id": "MDU6TGFiZWwxMjA1NzY1MDU0", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:macOS", "name": "subtype:macOS", "color": "b619ea", "default": false, "description": "macOS Build/Installation issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @yaddayaddayaddayadda ,\r\n\r\nI am able to import `keras_nlp` first followed by `tensorflow`.\r\n\r\nInput file 62443.py:\r\n\r\n```\r\nimport keras_nlp\r\nimport tensorflow as tf\r\n\r\nprint('keras_nlp version:',keras_nlp.__version__)\r\nprint('TF version:', tf.__version__)\r\n```\r\n\r\n\r\nOutput:\r\n```\r\n(tf-metal) suryanarayanay-macbookpro:Downloads suryanarayanay$ python 62443.py \r\nUsing TensorFlow backend\r\nkeras_nlp version: 0.6.3\r\nTF version: 2.15.0\r\n```\r\n\r\n", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62443\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62443\">No</a>\n" ]	2023-11-20T19:53:02	2023-11-22T05:56:38	2023-11-22T05:56:35	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version tensorflow-macos 2.14 ### Custom code Yes ### OS platform and distribution Mac OS Ventura 13.5.2 ### Mobile device _No response_ ### Python version 3.9.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Importing keras_nlp (version 0.6.3) before importing Tensorflw in a python package results in bus error on mac M1. I.e, this gives an error: import keras_nlp import tensorflow While this doesn't: import tensorflow import keras_nlp ### Standalone code to reproduce the issue ```shell import keras_nlp import tensorflow ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62443/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62443/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62442	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62442/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62442/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62442/events	https://github.com/tensorflow/tensorflow/issues/62442	2,002,689,242	I_kwDOArmXAs53Xpza	62,442	TF-OPT issue: StatelessRandomUniformV2 requires Compile Time Constant	{ "login": "monowaranjum", "id": 19803082, "node_id": "MDQ6VXNlcjE5ODAzMDgy", "avatar_url": "https://avatars.githubusercontent.com/u/19803082?v=4", "gravatar_id": "", "url": "https://api.github.com/users/monowaranjum", "html_url": "https://github.com/monowaranjum", "followers_url": "https://api.github.com/users/monowaranjum/followers", "following_url": "https://api.github.com/users/monowaranjum/following{/other_user}", "gists_url": "https://api.github.com/users/monowaranjum/gists{/gist_id}", "starred_url": "https://api.github.com/users/monowaranjum/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/monowaranjum/subscriptions", "organizations_url": "https://api.github.com/users/monowaranjum/orgs", "repos_url": "https://api.github.com/users/monowaranjum/repos", "events_url": "https://api.github.com/users/monowaranjum/events{/privacy}", "received_events_url": "https://api.github.com/users/monowaranjum/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1478826728, "node_id": "MDU6TGFiZWwxNDc4ODI2NzI4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:core", "name": "comp:core", "color": "024391", "default": false, "description": "issues related to core part of tensorflow" } ]	closed	false	{ "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false }	[ { "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false } ]	null	[ "@monowaranjum,\r\nI tried to execute the mentioned above on google colab with the latest stable tensorflow version and it was executed without any issue/error. Kindly find the gist of it [here](https://colab.research.google.com/gist/tilakrayal/e5a090e8bcc172e2145e200e6ae1534c/untitled1527.ipynb). Thank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62442\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62442\">No</a>\n" ]	2023-11-20T17:35:04	2023-11-29T15:59:21	2023-11-29T15:59:18	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.16.0-dev20231013 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04 ### Mobile device _No response_ ### Python version 3.10 ### Bazel version 5.1.1 ### GCC/compiler version g++-11 ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I built a custom model which classifies the mnist digits. Then used ```tf-mlir-translate``` to convert the saved model to tf dialect mlir. Then I tried to use the ```tf-opt``` with ```--tf-to-hlo-pipeline``` pass to get the hlo dialect from tensorflow dialect. It throws an error that says: ``` W tensorflow/core/framework/op_kernel.cc:1839] OP_REQUIRES failed at stateless_random_ops_v2.cc:93 : INVALID_ARGUMENT: Input 0 to node `tf.StatelessRandomUniformV2` with op StatelessRandomUniformV2 must be a compile-time constant. XLA compilation requires that operator arguments that represent shapes or dimensions be evaluated to concrete values at compile time. This error means that a shape or dimension argument could not be evaluated at compile time, usually because the value of the argument depends on a parameter to the computation, on a variable, or on a stateful operation such as a random number generator. /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:21:59: remark: compilation to HLO failed: INVALID_ARGUMENT: Input 0 to node `tf.StatelessRandomUniformV2` with op StatelessRandomUniformV2 must be a compile-time constant. XLA compilation requires that operator arguments that represent shapes or dimensions be evaluated to concrete values at compile time. This error means that a shape or dimension argument could not be evaluated at compile time, usually because the value of the argument depends on a parameter to the computation, on a variable, or on a stateful operation such as a random number generator. ``` ### Standalone code to reproduce the issue ```shell # Here is the model code: import os from tensorflow import mlir import tensorflow_hub as hub import tensorflow as tf import keras os.environ["TFHUB_DOWNLOAD_PROGRESS"] = "True" def get_model(): g = tf.Graph() with g.as_default(): tf.random.set_seed(42) model = keras.models.Sequential() model.add(keras.layers.Input((32,784))) model.add(keras.layers.Dense(64, activation='relu')) model.add(keras.layers.Dense(10, activation='softmax')) tf.io.write_graph(g.as_graph_def(), "logdir" , "mnist.pb") return model get_model() # Here is the command I ran for tf-mlir-translate ./tf-mlir-translate --graphdef-to-mlir --tf-enable-shape-inference-on-import \ /home/rashik/Documents/tensorfow_mlir_test/mnist.pb -o \ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir # Here is the command I ran for tf-opt ./tf-opt --tf-to-hlo-pipeline /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir -o /home/rashik/Documents/tensorfow_mlir_test/hlo_dialect.mlir ``` ### Relevant log output ```shell // Output from tf-mlir-translate: module attributes {tf.versions = {bad_consumers = [], min_consumer = 0 : i32, producer = 1647 : i32}} { func.func @main() { tf_executor.graph { %outputs, %control = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense/bias"}> {_class = ["loc:@dense/bias"], allowed_devices = [], debug_name = "dense/bias/", device = ""} : () -> tensor<!tf_type.resource<tensor<64xf32>>> %outputs_0, %control_1 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<64xf32> %outputs_2, %control_3 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<i1> %outputs_4, %control_5 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<64xf32> %outputs_6, %control_7 = tf_executor.island wraps "tf.Const"() <{value = dense<0.000000e+00> : tensor<64xf32>}> {_class = ["loc:@dense/bias"], device = ""} : () -> tensor<64xf32> %control_8 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs, %outputs_6) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>, tensor<64xf32>) -> () %outputs_9, %control_10 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense/kernel"}> {_class = ["loc:@dense/kernel"], allowed_devices = [], debug_name = "dense/kernel/", device = ""} : () -> tensor<!tf_type.resource<tensor<784x64xf32>>> %outputs_11, %control_12 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<784x64xf32> %outputs_13, %control_14 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<i1> %outputs_15, %control_16 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<784x64xf32> %outputs_17, %control_18 = tf_executor.island wraps "tf.Const"() <{value = dense<[835938099, 0]> : tensor<2xi32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<2xi32> %outputs_19:2, %control_20 = tf_executor.island wraps "tf.StatelessRandomGetKeyCounter"(%outputs_17) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<2xi32>) -> (tensor<1xui64>, tensor<2xui64>) %outputs_21, %control_22 = tf_executor.island wraps "tf.Const"() <{value = dense<3> : tensor<i32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<i32> %outputs_23, %control_24 = tf_executor.island wraps "tf.Const"() <{value = dense<0.0841158255> : tensor<f32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<f32> %outputs_25, %control_26 = tf_executor.island wraps "tf.Const"() <{value = dense<-0.0841158255> : tensor<f32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<f32> %outputs_27, %control_28 = tf_executor.island wraps "tf.Sub"(%outputs_23, %outputs_25) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<f32>, tensor<f32>) -> tensor<f32> %outputs_29, %control_30 = tf_executor.island wraps "tf.Const"() <{value = dense<[784, 64]> : tensor<2xi32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<2xi32> %outputs_31, %control_32 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_29, %outputs_19#0, %outputs_19#1, %outputs_21) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<784x64xf32> %outputs_33, %control_34 = tf_executor.island wraps "tf.Mul"(%outputs_31, %outputs_27) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<784x64xf32>, tensor<f32>) -> tensor<784x64xf32> %outputs_35, %control_36 = tf_executor.island wraps "tf.AddV2"(%outputs_33, %outputs_25) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<784x64xf32>, tensor<f32>) -> tensor<784x64xf32> %control_37 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_9, %outputs_35) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>, tensor<784x64xf32>) -> () %outputs_38, %control_39 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense_1/bias"}> {_class = ["loc:@dense_1/bias"], allowed_devices = [], debug_name = "dense_1/bias/", device = ""} : () -> tensor<!tf_type.resource<tensor<10xf32>>> %outputs_40, %control_41 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<10xf32> %outputs_42, %control_43 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<i1> %outputs_44, %control_45 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<10xf32> %outputs_46, %control_47 = tf_executor.island wraps "tf.Const"() <{value = dense<0.000000e+00> : tensor<10xf32>}> {_class = ["loc:@dense_1/bias"], device = ""} : () -> tensor<10xf32> %control_48 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_38, %outputs_46) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>, tensor<10xf32>) -> () %outputs_49, %control_50 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense_1/kernel"}> {_class = ["loc:@dense_1/kernel"], allowed_devices = [], debug_name = "dense_1/kernel/", device = ""} : () -> tensor<!tf_type.resource<tensor<64x10xf32>>> %outputs_51, %control_52 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<64x10xf32> %outputs_53, %control_54 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<i1> %outputs_55, %control_56 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<64x10xf32> %outputs_57, %control_58 = tf_executor.island wraps "tf.Const"() <{value = dense<[9356290, 0]> : tensor<2xi32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<2xi32> %outputs_59:2, %control_60 = tf_executor.island wraps "tf.StatelessRandomGetKeyCounter"(%outputs_57) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<2xi32>) -> (tensor<1xui64>, tensor<2xui64>) %outputs_61, %control_62 = tf_executor.island wraps "tf.Const"() <{value = dense<3> : tensor<i32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<i32> %outputs_63, %control_64 = tf_executor.island wraps "tf.Const"() <{value = dense<0.284747392> : tensor<f32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<f32> %outputs_65, %control_66 = tf_executor.island wraps "tf.Const"() <{value = dense<-0.284747392> : tensor<f32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<f32> %outputs_67, %control_68 = tf_executor.island wraps "tf.Sub"(%outputs_63, %outputs_65) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<f32>, tensor<f32>) -> tensor<f32> %outputs_69, %control_70 = tf_executor.island wraps "tf.Const"() <{value = dense<[64, 10]> : tensor<2xi32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<2xi32> %outputs_71, %control_72 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_69, %outputs_59#0, %outputs_59#1, %outputs_61) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<64x10xf32> %outputs_73, %control_74 = tf_executor.island wraps "tf.Mul"(%outputs_71, %outputs_67) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<64x10xf32>, tensor<f32>) -> tensor<64x10xf32> %outputs_75, %control_76 = tf_executor.island wraps "tf.AddV2"(%outputs_73, %outputs_65) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<64x10xf32>, tensor<f32>) -> tensor<64x10xf32> %control_77 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_49, %outputs_75) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>, tensor<64x10xf32>) -> () %outputs_78, %control_79 = tf_executor.island wraps "tf.Placeholder"() {device = "", shape = #tf_type.shape<?x784>} : () -> tensor<?x784xf32> %outputs_80, %control_81 = tf_executor.island wraps "tf.MatMul"(%outputs_78, %outputs_11) <{transpose_a = false, transpose_b = false}> {device = ""} : (tensor<?x784xf32>, tensor<784x64xf32>) -> tensor<?x64xf32> %outputs_82, %control_83 = tf_executor.island wraps "tf.BiasAdd"(%outputs_80, %outputs_0) <{data_format = "NHWC"}> {device = ""} : (tensor<?x64xf32>, tensor<64xf32>) -> tensor<?x64xf32> %outputs_84, %control_85 = tf_executor.island wraps "tf.Relu"(%outputs_82) {device = ""} : (tensor<?x64xf32>) -> tensor<?x64xf32> %outputs_86, %control_87 = tf_executor.island wraps "tf.MatMul"(%outputs_84, %outputs_51) <{transpose_a = false, transpose_b = false}> {device = ""} : (tensor<?x64xf32>, tensor<64x10xf32>) -> tensor<?x10xf32> %outputs_88, %control_89 = tf_executor.island wraps "tf.BiasAdd"(%outputs_86, %outputs_40) <{data_format = "NHWC"}> {device = ""} : (tensor<?x10xf32>, tensor<10xf32>) -> tensor<?x10xf32> %outputs_90, %control_91 = tf_executor.island wraps "tf.Softmax"(%outputs_88) {device = ""} : (tensor<?x10xf32>) -> tensor<?x10xf32> tf_executor.fetch } return } } *********************** Log Output ********************** 2023-11-20 11:15:08.462092: W tensorflow/core/framework/op_kernel.cc:1839] OP_REQUIRES failed at stateless_random_ops_v2.cc:93 : INVALID_ARGUMENT: Input 0 to node `tf.StatelessRandomUniformV2` with op StatelessRandomUniformV2 must be a compile-time constant. XLA compilation requires that operator arguments that represent shapes or dimensions be evaluated to concrete values at compile time. This error means that a shape or dimension argument could not be evaluated at compile time, usually because the value of the argument depends on a parameter to the computation, on a variable, or on a stateful operation such as a random number generator. /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:21:59: remark: compilation to HLO failed: INVALID_ARGUMENT: Input 0 to node `tf.StatelessRandomUniformV2` with op StatelessRandomUniformV2 must be a compile-time constant. XLA compilation requires that operator arguments that represent shapes or dimensions be evaluated to concrete values at compile time. This error means that a shape or dimension argument could not be evaluated at compile time, usually because the value of the argument depends on a parameter to the computation, on a variable, or on a stateful operation such as a random number generator. %outputs_31, %control_32 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_29, %outputs_19#0, %outputs_19#1, %outputs_21) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:21:59: note: see current operation: %0 = "tf.StatelessRandomUniformV2"(%outputs_29, %outputs_19#0, %outputs_19#1, %outputs_21) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<784x64xf32> 2023-11-20 11:15:08.463988: W tensorflow/core/framework/op_kernel.cc:1839] OP_REQUIRES failed at stateless_random_ops_v2.cc:93 : INVALID_ARGUMENT: Input 0 to node `tf.StatelessRandomUniformV2` with op StatelessRandomUniformV2 must be a compile-time constant. XLA compilation requires that operator arguments that represent shapes or dimensions be evaluated to concrete values at compile time. This error means that a shape or dimension argument could not be evaluated at compile time, usually because the value of the argument depends on a parameter to the computation, on a variable, or on a stateful operation such as a random number generator. /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:42:59: remark: compilation to HLO failed: INVALID_ARGUMENT: Input 0 to node `tf.StatelessRandomUniformV2` with op StatelessRandomUniformV2 must be a compile-time constant. XLA compilation requires that operator arguments that represent shapes or dimensions be evaluated to concrete values at compile time. This error means that a shape or dimension argument could not be evaluated at compile time, usually because the value of the argument depends on a parameter to the computation, on a variable, or on a stateful operation such as a random number generator. %outputs_71, %control_72 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_69, %outputs_59#0, %outputs_59#1, %outputs_61) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:42:59: note: see current operation: %0 = "tf.StatelessRandomUniformV2"(%outputs_69, %outputs_59#0, %outputs_59#1, %outputs_61) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<64x10xf32> /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:4:53: error: Could not legalize op: tf.VarHandleOp %outputs, %control = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense/bias"}> {_class = ["loc:@dense/bias"], allowed_devices = [], debug_name = "dense/bias/", device = ""} : () -> tensor<!tf_type.resource<tensor<64xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:4:53: error: Could not legalize op: tf_executor.yield %outputs, %control = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense/bias"}> {_class = ["loc:@dense/bias"], allowed_devices = [], debug_name = "dense/bias/", device = ""} : () -> tensor<!tf_type.resource<tensor<64xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:4:53: error: Could not legalize op: tf_executor.island %outputs, %control = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense/bias"}> {_class = ["loc:@dense/bias"], allowed_devices = [], debug_name = "dense/bias/", device = ""} : () -> tensor<!tf_type.resource<tensor<64xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:5:57: error: Could not legalize op: tf.ReadVariableOp %outputs_0, %control_1 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:5:57: error: Could not legalize op: tf_executor.yield %outputs_0, %control_1 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:5:57: error: Could not legalize op: tf_executor.island %outputs_0, %control_1 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:6:57: error: Could not legalize op: tf.VarIsInitializedOp %outputs_2, %control_3 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:6:57: error: Could not legalize op: tf_executor.yield %outputs_2, %control_3 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:6:57: error: Could not legalize op: tf_executor.island %outputs_2, %control_3 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:7:57: error: Could not legalize op: tf.ReadVariableOp %outputs_4, %control_5 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:7:57: error: Could not legalize op: tf_executor.yield %outputs_4, %control_5 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:7:57: error: Could not legalize op: tf_executor.island %outputs_4, %control_5 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs) {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>) -> tensor<64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:8:57: error: Could not legalize op: tf_executor.yield %outputs_6, %control_7 = tf_executor.island wraps "tf.Const"() <{value = dense<0.000000e+00> : tensor<64xf32>}> {_class = ["loc:@dense/bias"], device = ""} : () -> tensor<64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:8:57: error: Could not legalize op: tf_executor.island %outputs_6, %control_7 = tf_executor.island wraps "tf.Const"() <{value = dense<0.000000e+00> : tensor<64xf32>}> {_class = ["loc:@dense/bias"], device = ""} : () -> tensor<64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:9:45: error: Could not legalize op: tf.AssignVariableOp %control_8 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs, %outputs_6) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>, tensor<64xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:9:45: error: Could not legalize op: tf_executor.yield %control_8 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs, %outputs_6) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>, tensor<64xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:9:45: error: Could not legalize op: tf_executor.island %control_8 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs, %outputs_6) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<64xf32>>>, tensor<64xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:10:58: error: Could not legalize op: tf.VarHandleOp %outputs_9, %control_10 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense/kernel"}> {_class = ["loc:@dense/kernel"], allowed_devices = [], debug_name = "dense/kernel/", device = ""} : () -> tensor<!tf_type.resource<tensor<784x64xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:10:58: error: Could not legalize op: tf_executor.yield %outputs_9, %control_10 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense/kernel"}> {_class = ["loc:@dense/kernel"], allowed_devices = [], debug_name = "dense/kernel/", device = ""} : () -> tensor<!tf_type.resource<tensor<784x64xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:10:58: error: Could not legalize op: tf_executor.island %outputs_9, %control_10 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense/kernel"}> {_class = ["loc:@dense/kernel"], allowed_devices = [], debug_name = "dense/kernel/", device = ""} : () -> tensor<!tf_type.resource<tensor<784x64xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:11:59: error: Could not legalize op: tf.ReadVariableOp %outputs_11, %control_12 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:11:59: error: Could not legalize op: tf_executor.yield %outputs_11, %control_12 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:11:59: error: Could not legalize op: tf_executor.island %outputs_11, %control_12 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:12:59: error: Could not legalize op: tf.VarIsInitializedOp %outputs_13, %control_14 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:12:59: error: Could not legalize op: tf_executor.yield %outputs_13, %control_14 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:12:59: error: Could not legalize op: tf_executor.island %outputs_13, %control_14 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:13:59: error: Could not legalize op: tf.ReadVariableOp %outputs_15, %control_16 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:13:59: error: Could not legalize op: tf_executor.yield %outputs_15, %control_16 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:13:59: error: Could not legalize op: tf_executor.island %outputs_15, %control_16 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_9) {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:14:59: error: Could not legalize op: tf_executor.yield %outputs_17, %control_18 = tf_executor.island wraps "tf.Const"() <{value = dense<[835938099, 0]> : tensor<2xi32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<2xi32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:14:59: error: Could not legalize op: tf_executor.island %outputs_17, %control_18 = tf_executor.island wraps "tf.Const"() <{value = dense<[835938099, 0]> : tensor<2xi32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<2xi32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:15:61: error: Could not legalize op: tf_executor.yield %outputs_19:2, %control_20 = tf_executor.island wraps "tf.StatelessRandomGetKeyCounter"(%outputs_17) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<2xi32>) -> (tensor<1xui64>, tensor<2xui64>) ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:15:61: error: Could not legalize op: tf_executor.island %outputs_19:2, %control_20 = tf_executor.island wraps "tf.StatelessRandomGetKeyCounter"(%outputs_17) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<2xi32>) -> (tensor<1xui64>, tensor<2xui64>) ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:16:59: error: Could not legalize op: tf_executor.yield %outputs_21, %control_22 = tf_executor.island wraps "tf.Const"() <{value = dense<3> : tensor<i32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<i32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:16:59: error: Could not legalize op: tf_executor.island %outputs_21, %control_22 = tf_executor.island wraps "tf.Const"() <{value = dense<3> : tensor<i32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<i32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:17:59: error: Could not legalize op: tf_executor.yield %outputs_23, %control_24 = tf_executor.island wraps "tf.Const"() <{value = dense<0.0841158255> : tensor<f32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:17:59: error: Could not legalize op: tf_executor.island %outputs_23, %control_24 = tf_executor.island wraps "tf.Const"() <{value = dense<0.0841158255> : tensor<f32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:18:59: error: Could not legalize op: tf_executor.yield %outputs_25, %control_26 = tf_executor.island wraps "tf.Const"() <{value = dense<-0.0841158255> : tensor<f32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:18:59: error: Could not legalize op: tf_executor.island %outputs_25, %control_26 = tf_executor.island wraps "tf.Const"() <{value = dense<-0.0841158255> : tensor<f32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:19:59: error: Could not legalize op: tf_executor.yield %outputs_27, %control_28 = tf_executor.island wraps "tf.Sub"(%outputs_23, %outputs_25) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<f32>, tensor<f32>) -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:19:59: error: Could not legalize op: tf_executor.island %outputs_27, %control_28 = tf_executor.island wraps "tf.Sub"(%outputs_23, %outputs_25) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<f32>, tensor<f32>) -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:20:59: error: Could not legalize op: tf_executor.yield %outputs_29, %control_30 = tf_executor.island wraps "tf.Const"() <{value = dense<[784, 64]> : tensor<2xi32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<2xi32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:20:59: error: Could not legalize op: tf_executor.island %outputs_29, %control_30 = tf_executor.island wraps "tf.Const"() <{value = dense<[784, 64]> : tensor<2xi32>}> {_class = ["loc:@dense/kernel"], device = ""} : () -> tensor<2xi32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:21:59: error: Could not legalize op: tf.StatelessRandomUniformV2 %outputs_31, %control_32 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_29, %outputs_19#0, %outputs_19#1, %outputs_21) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:21:59: error: Could not legalize op: tf_executor.yield %outputs_31, %control_32 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_29, %outputs_19#0, %outputs_19#1, %outputs_21) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:21:59: error: Could not legalize op: tf_executor.island %outputs_31, %control_32 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_29, %outputs_19#0, %outputs_19#1, %outputs_21) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:22:59: error: Could not legalize op: tf_executor.yield %outputs_33, %control_34 = tf_executor.island wraps "tf.Mul"(%outputs_31, %outputs_27) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<784x64xf32>, tensor<f32>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:22:59: error: Could not legalize op: tf_executor.island %outputs_33, %control_34 = tf_executor.island wraps "tf.Mul"(%outputs_31, %outputs_27) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<784x64xf32>, tensor<f32>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:23:59: error: Could not legalize op: tf_executor.yield %outputs_35, %control_36 = tf_executor.island wraps "tf.AddV2"(%outputs_33, %outputs_25) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<784x64xf32>, tensor<f32>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:23:59: error: Could not legalize op: tf_executor.island %outputs_35, %control_36 = tf_executor.island wraps "tf.AddV2"(%outputs_33, %outputs_25) {_class = ["loc:@dense/kernel"], device = ""} : (tensor<784x64xf32>, tensor<f32>) -> tensor<784x64xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:24:46: error: Could not legalize op: tf.AssignVariableOp %control_37 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_9, %outputs_35) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>, tensor<784x64xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:24:46: error: Could not legalize op: tf_executor.yield %control_37 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_9, %outputs_35) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>, tensor<784x64xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:24:46: error: Could not legalize op: tf_executor.island %control_37 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_9, %outputs_35) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<784x64xf32>>>, tensor<784x64xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:25:59: error: Could not legalize op: tf.VarHandleOp %outputs_38, %control_39 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense_1/bias"}> {_class = ["loc:@dense_1/bias"], allowed_devices = [], debug_name = "dense_1/bias/", device = ""} : () -> tensor<!tf_type.resource<tensor<10xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:25:59: error: Could not legalize op: tf_executor.yield %outputs_38, %control_39 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense_1/bias"}> {_class = ["loc:@dense_1/bias"], allowed_devices = [], debug_name = "dense_1/bias/", device = ""} : () -> tensor<!tf_type.resource<tensor<10xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:25:59: error: Could not legalize op: tf_executor.island %outputs_38, %control_39 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense_1/bias"}> {_class = ["loc:@dense_1/bias"], allowed_devices = [], debug_name = "dense_1/bias/", device = ""} : () -> tensor<!tf_type.resource<tensor<10xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:26:59: error: Could not legalize op: tf.ReadVariableOp %outputs_40, %control_41 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:26:59: error: Could not legalize op: tf_executor.yield %outputs_40, %control_41 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:26:59: error: Could not legalize op: tf_executor.island %outputs_40, %control_41 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:27:59: error: Could not legalize op: tf.VarIsInitializedOp %outputs_42, %control_43 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:27:59: error: Could not legalize op: tf_executor.yield %outputs_42, %control_43 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:27:59: error: Could not legalize op: tf_executor.island %outputs_42, %control_43 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:28:59: error: Could not legalize op: tf.ReadVariableOp %outputs_44, %control_45 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:28:59: error: Could not legalize op: tf_executor.yield %outputs_44, %control_45 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:28:59: error: Could not legalize op: tf_executor.island %outputs_44, %control_45 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_38) {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>) -> tensor<10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:29:59: error: Could not legalize op: tf_executor.yield %outputs_46, %control_47 = tf_executor.island wraps "tf.Const"() <{value = dense<0.000000e+00> : tensor<10xf32>}> {_class = ["loc:@dense_1/bias"], device = ""} : () -> tensor<10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:29:59: error: Could not legalize op: tf_executor.island %outputs_46, %control_47 = tf_executor.island wraps "tf.Const"() <{value = dense<0.000000e+00> : tensor<10xf32>}> {_class = ["loc:@dense_1/bias"], device = ""} : () -> tensor<10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:30:46: error: Could not legalize op: tf.AssignVariableOp %control_48 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_38, %outputs_46) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>, tensor<10xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:30:46: error: Could not legalize op: tf_executor.yield %control_48 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_38, %outputs_46) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>, tensor<10xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:30:46: error: Could not legalize op: tf_executor.island %control_48 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_38, %outputs_46) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<10xf32>>>, tensor<10xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:31:59: error: Could not legalize op: tf.VarHandleOp %outputs_49, %control_50 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense_1/kernel"}> {_class = ["loc:@dense_1/kernel"], allowed_devices = [], debug_name = "dense_1/kernel/", device = ""} : () -> tensor<!tf_type.resource<tensor<64x10xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:31:59: error: Could not legalize op: tf_executor.yield %outputs_49, %control_50 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense_1/kernel"}> {_class = ["loc:@dense_1/kernel"], allowed_devices = [], debug_name = "dense_1/kernel/", device = ""} : () -> tensor<!tf_type.resource<tensor<64x10xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:31:59: error: Could not legalize op: tf_executor.island %outputs_49, %control_50 = tf_executor.island wraps "tf.VarHandleOp"() <{container = "", shared_name = "dense_1/kernel"}> {_class = ["loc:@dense_1/kernel"], allowed_devices = [], debug_name = "dense_1/kernel/", device = ""} : () -> tensor<!tf_type.resource<tensor<64x10xf32>>> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:32:59: error: Could not legalize op: tf.ReadVariableOp %outputs_51, %control_52 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:32:59: error: Could not legalize op: tf_executor.yield %outputs_51, %control_52 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:32:59: error: Could not legalize op: tf_executor.island %outputs_51, %control_52 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:33:59: error: Could not legalize op: tf.VarIsInitializedOp %outputs_53, %control_54 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:33:59: error: Could not legalize op: tf_executor.yield %outputs_53, %control_54 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:33:59: error: Could not legalize op: tf_executor.island %outputs_53, %control_54 = tf_executor.island wraps "tf.VarIsInitializedOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<i1> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:34:59: error: Could not legalize op: tf.ReadVariableOp %outputs_55, %control_56 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:34:59: error: Could not legalize op: tf_executor.yield %outputs_55, %control_56 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:34:59: error: Could not legalize op: tf_executor.island %outputs_55, %control_56 = tf_executor.island wraps "tf.ReadVariableOp"(%outputs_49) {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:35:59: error: Could not legalize op: tf_executor.yield %outputs_57, %control_58 = tf_executor.island wraps "tf.Const"() <{value = dense<[9356290, 0]> : tensor<2xi32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<2xi32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:35:59: error: Could not legalize op: tf_executor.island %outputs_57, %control_58 = tf_executor.island wraps "tf.Const"() <{value = dense<[9356290, 0]> : tensor<2xi32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<2xi32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:36:61: error: Could not legalize op: tf_executor.yield %outputs_59:2, %control_60 = tf_executor.island wraps "tf.StatelessRandomGetKeyCounter"(%outputs_57) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<2xi32>) -> (tensor<1xui64>, tensor<2xui64>) ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:36:61: error: Could not legalize op: tf_executor.island %outputs_59:2, %control_60 = tf_executor.island wraps "tf.StatelessRandomGetKeyCounter"(%outputs_57) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<2xi32>) -> (tensor<1xui64>, tensor<2xui64>) ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:37:59: error: Could not legalize op: tf_executor.yield %outputs_61, %control_62 = tf_executor.island wraps "tf.Const"() <{value = dense<3> : tensor<i32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<i32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:37:59: error: Could not legalize op: tf_executor.island %outputs_61, %control_62 = tf_executor.island wraps "tf.Const"() <{value = dense<3> : tensor<i32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<i32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:38:59: error: Could not legalize op: tf_executor.yield %outputs_63, %control_64 = tf_executor.island wraps "tf.Const"() <{value = dense<0.284747392> : tensor<f32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:38:59: error: Could not legalize op: tf_executor.island %outputs_63, %control_64 = tf_executor.island wraps "tf.Const"() <{value = dense<0.284747392> : tensor<f32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:39:59: error: Could not legalize op: tf_executor.yield %outputs_65, %control_66 = tf_executor.island wraps "tf.Const"() <{value = dense<-0.284747392> : tensor<f32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:39:59: error: Could not legalize op: tf_executor.island %outputs_65, %control_66 = tf_executor.island wraps "tf.Const"() <{value = dense<-0.284747392> : tensor<f32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:40:59: error: Could not legalize op: tf_executor.yield %outputs_67, %control_68 = tf_executor.island wraps "tf.Sub"(%outputs_63, %outputs_65) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<f32>, tensor<f32>) -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:40:59: error: Could not legalize op: tf_executor.island %outputs_67, %control_68 = tf_executor.island wraps "tf.Sub"(%outputs_63, %outputs_65) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<f32>, tensor<f32>) -> tensor<f32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:41:59: error: Could not legalize op: tf_executor.yield %outputs_69, %control_70 = tf_executor.island wraps "tf.Const"() <{value = dense<[64, 10]> : tensor<2xi32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<2xi32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:41:59: error: Could not legalize op: tf_executor.island %outputs_69, %control_70 = tf_executor.island wraps "tf.Const"() <{value = dense<[64, 10]> : tensor<2xi32>}> {_class = ["loc:@dense_1/kernel"], device = ""} : () -> tensor<2xi32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:42:59: error: Could not legalize op: tf.StatelessRandomUniformV2 %outputs_71, %control_72 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_69, %outputs_59#0, %outputs_59#1, %outputs_61) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:42:59: error: Could not legalize op: tf_executor.yield %outputs_71, %control_72 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_69, %outputs_59#0, %outputs_59#1, %outputs_61) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:42:59: error: Could not legalize op: tf_executor.island %outputs_71, %control_72 = tf_executor.island wraps "tf.StatelessRandomUniformV2"(%outputs_69, %outputs_59#0, %outputs_59#1, %outputs_61) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<2xi32>, tensor<1xui64>, tensor<2xui64>, tensor<i32>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:43:59: error: Could not legalize op: tf_executor.yield %outputs_73, %control_74 = tf_executor.island wraps "tf.Mul"(%outputs_71, %outputs_67) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<64x10xf32>, tensor<f32>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:43:59: error: Could not legalize op: tf_executor.island %outputs_73, %control_74 = tf_executor.island wraps "tf.Mul"(%outputs_71, %outputs_67) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<64x10xf32>, tensor<f32>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:44:59: error: Could not legalize op: tf_executor.yield %outputs_75, %control_76 = tf_executor.island wraps "tf.AddV2"(%outputs_73, %outputs_65) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<64x10xf32>, tensor<f32>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:44:59: error: Could not legalize op: tf_executor.island %outputs_75, %control_76 = tf_executor.island wraps "tf.AddV2"(%outputs_73, %outputs_65) {_class = ["loc:@dense_1/kernel"], device = ""} : (tensor<64x10xf32>, tensor<f32>) -> tensor<64x10xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:45:46: error: Could not legalize op: tf.AssignVariableOp %control_77 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_49, %outputs_75) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>, tensor<64x10xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:45:46: error: Could not legalize op: tf_executor.yield %control_77 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_49, %outputs_75) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>, tensor<64x10xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:45:46: error: Could not legalize op: tf_executor.island %control_77 = tf_executor.island wraps "tf.AssignVariableOp"(%outputs_49, %outputs_75) <{validate_shape = false}> {device = ""} : (tensor<!tf_type.resource<tensor<64x10xf32>>>, tensor<64x10xf32>) -> () ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:46:59: error: Could not legalize op: tf.Placeholder %outputs_78, %control_79 = tf_executor.island wraps "tf.Placeholder"() {device = "", shape = #tf_type.shape<?x784>} : () -> tensor<?x784xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:46:59: error: Could not legalize op: tf_executor.yield %outputs_78, %control_79 = tf_executor.island wraps "tf.Placeholder"() {device = "", shape = #tf_type.shape<?x784>} : () -> tensor<?x784xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:46:59: error: Could not legalize op: tf_executor.island %outputs_78, %control_79 = tf_executor.island wraps "tf.Placeholder"() {device = "", shape = #tf_type.shape<?x784>} : () -> tensor<?x784xf32> ^ /home/rashik/Documents/tensorfow_mlir_test/tf_dialect.mlir:47:59: error: Node `mhlo.dot` must have compile-time constant inputs and outputs. XLA compilation requires that operator arguments that represent shapes or dimensions be evaluated to concrete values at compile time. This error means that a shape or dimension argument could not be evaluated at compile time, usually because the value of the argument depends on a parameter to the computation, on a variable, or on a stateful operation such as a random number generator. %outputs_80, %control_81 = tf_executor.island wraps "tf.MatMul"(%outputs_78, %outputs_11) <{transpose_a = false, transpose_b = false}> {device = ""} : (tensor<?x784xf32>, tensor<784x64xf32>) -> tensor<?x64xf32> ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62442/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62442/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62441	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62441/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62441/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62441/events	https://github.com/tensorflow/tensorflow/issues/62441	2,002,503,931	I_kwDOArmXAs53W8j7	62,441	Convolution: CPU memory increase with growing number of different sequence lengths	{ "login": "vieting", "id": 45091115, "node_id": "MDQ6VXNlcjQ1MDkxMTE1", "avatar_url": "https://avatars.githubusercontent.com/u/45091115?v=4", "gravatar_id": "", "url": "https://api.github.com/users/vieting", "html_url": "https://github.com/vieting", "followers_url": "https://api.github.com/users/vieting/followers", "following_url": "https://api.github.com/users/vieting/following{/other_user}", "gists_url": "https://api.github.com/users/vieting/gists{/gist_id}", "starred_url": "https://api.github.com/users/vieting/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/vieting/subscriptions", "organizations_url": "https://api.github.com/users/vieting/orgs", "repos_url": "https://api.github.com/users/vieting/repos", "events_url": "https://api.github.com/users/vieting/events{/privacy}", "received_events_url": "https://api.github.com/users/vieting/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1463677878, "node_id": "MDU6TGFiZWwxNDYzNjc3ODc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:performance", "name": "type:performance", "color": "159b2e", "default": false, "description": "Performance Issue" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "While searching for related issues, we found these PyTorch issues: https://github.com/pytorch/pytorch/issues/98688 https://github.com/pytorch/pytorch/issues/101921\r\n\r\nSpecifically:\r\n\r\n> this is likely a consequence of cuDNN V8 API execution plans taking up more host memory than algo perf data did in V7. Some additional memory usage is expected with each dynamic shape as these execution plans are cached in case the same workload is reached\r\n", "Indeed, the behavior is very similar to what was described in PyTorch version 2.0.0+cu117 as used in https://github.com/pytorch/pytorch/issues/98688. With PyTorch 2.1.0+cu121, the issue seems to be solved.\r\n\r\n![image](https://github.com/tensorflow/tensorflow/assets/45091115/fedbdb03-7cd9-4f49-8044-8943830dba8a)", "I created a colab which reproduces one of the curves in the above figures. See [here](https://colab.research.google.com/drive/1whYaC3ZgSUr86z44pkUOVLQikMu_22sw?usp=sharing)." ]	2023-11-20T15:51:19	2023-12-06T00:26:21	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution Linux Ubuntu 22.04 ### Mobile device _No response_ ### Python version 3.11 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version CUDA V11.8.89, cuDNN version 8600 ### GPU model and memory NVIDIA GeForce GTX 1080 Ti ### Current behavior? I noticed a linear increase of CPU memory usage in my setups when using a convolution on raw waveforms (i.e., sequences which are long in time and 1D in feature). I could isolate the issue and it seems to be related to the number of different sequence lengths that occur. I.e., if the sequence length is fixed to 100k, the memory consumption is constant. If it is randomly sampled from a given range, the memory consumption asymptotically grows towards a larger value as the range gets larger. This can be observed in the plot below. Also note that the memory consumption is not influenced by the absolute sequence length, just by the size of the range. ![image](https://github.com/tensorflow/tensorflow/assets/45091115/cbaeaadf-6d95-434f-bf2f-1a4aec232efc) I measured the memory consumption using `watch_memory()` from [here](https://github.com/rwth-i6/returnn/blob/c230d1408a2d7620a9d00a5171c998d3876e69dc/returnn/util/watch_memory.py#L13). The different runs in the plot correspond to different `n_time_min` and `n_time_max` in the stand-alone code. I reproduced the issue with an apptainer image built on top of the [tensorflow 2.14 image from dockerhub](https://hub.docker.com/layers/tensorflow/tensorflow/2.14.0-gpu-jupyter/images/sha256-981372796921ef7bb75f4fe5fbe98c335824d08233bed57586633199028d5e18?context=explore). The image definition file looks as follows: <details> ``` Bootstrap: docker From: tensorflow/tensorflow:2.14.0-gpu Stage: build %post apt update -y # all the fundamental basics, zsh is need because calling the cache manager might launch the user shell DEBIAN_FRONTEND=noninteractive apt install -y wget git unzip gzip libssl-dev lsb-release zsh \ bison libxml2-dev libopenblas-dev libsndfile1-dev libcrypto++-dev libcppunit-dev \ parallel xmlstarlet python3-lxml htop strace gdb sox python3-pip cmake ffmpeg vim cd /usr/local git clone https://github.com/rwth-i6/cache-manager.git cd bin ln -s ../cache-manager/cf cf echo /usr/local/lib/python3.11/dist-packages/tensorflow > /etc/ld.so.conf.d/tensorflow.conf ldconfig apt install -y python3 python3-pip pip3 install -U pip setuptools wheel pip3 install ipdb pip3 install h5py six soundfile librosa==0.10 better-exchook dm-tree psutil pip3 install --ignore-installed psutil flask ipython pip3 install git+https://github.com/rwth-i6/sisyphus pip3 install black==22.3.0 matplotlib typing-extensions typeguard # sequitur-g2p==1.0.1668.23 pip3 install memray objgraph Pympler ``` </details> ### Standalone code to reproduce the issue ```shell import numpy as np import tensorflow as tf n_feat = 1 n_out = 30 filter_size = 5 n_steps = 100000 n_time_min = 10000 n_time_max = 30000 batch_size_max = 400000 filters = tf.Variable(tf.random.normal((filter_size, n_feat, n_out), stddev=0.01)) for step in range(n_steps): n_time = np.random.randint(n_time_min, n_time_max) n_batch = batch_size_max // n_time x = tf.random.normal((n_batch, n_time, n_feat)) y = tf.nn.convolution( x, filters=filters, padding="VALID", ) ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62441/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62441/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62440	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62440/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62440/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62440/events	https://github.com/tensorflow/tensorflow/issues/62440	2,002,019,042	I_kwDOArmXAs53VGLi	62,440	Bug in memory out-of-bound related to kMaxDim of RuntimeShape at tflite kernels	{ "login": "Seunghui98", "id": 54658745, "node_id": "MDQ6VXNlcjU0NjU4NzQ1", "avatar_url": "https://avatars.githubusercontent.com/u/54658745?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Seunghui98", "html_url": "https://github.com/Seunghui98", "followers_url": "https://api.github.com/users/Seunghui98/followers", "following_url": "https://api.github.com/users/Seunghui98/following{/other_user}", "gists_url": "https://api.github.com/users/Seunghui98/gists{/gist_id}", "starred_url": "https://api.github.com/users/Seunghui98/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Seunghui98/subscriptions", "organizations_url": "https://api.github.com/users/Seunghui98/orgs", "repos_url": "https://api.github.com/users/Seunghui98/repos", "events_url": "https://api.github.com/users/Seunghui98/events{/privacy}", "received_events_url": "https://api.github.com/users/Seunghui98/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 3797168204, "node_id": "LA_kwDOArmXAs7iVDBM", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.8", "name": "TF 2.8", "color": "5DC9D0", "default": false, "description": "" } ]	closed	false	{ "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false }	[ { "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false } ]	null	[ "Hmmm.. I'm not sure but the error occurred at the following codes.\r\n```c++\r\n// tensorflow/lite/kernels/internal/runtime_shape.h L179-L180\r\nstd::memcpy(DimsData() + size_increase, shape.DimsData(),\r\n sizeof(int32_t) * shape.DimensionsCount());\r\n```\r\n\r\nI think there could be three candidates that cause an error.\r\n1. Caller fault\r\n2. Callee fault\r\n3. Compiler bug\r\n\r\nWhen it comes to the third one, I think the compiler doesn't evaluate properly `DimsData` which should return `dims_` or `dims_pointer` according to its size. Well, but I don't know exact reason. I just show you trial and error codes where the build is successful with that codes, which might help you guess what causes an errorr.\r\n\r\n1. Just change `kMaxSmallSize` to 8 instead of 5.\r\n4. Make `dims_` a pointer.\r\n\r\n<details>\r\n<summary> diff </summary>\r\n\r\n```diff\r\ndiff --git a/tensorflow/lite/kernels/internal/runtime_shape.h b/tensorflow/lite/kernels/internal/runtime_shape.h\r\nindex 855c305d270..3c1e0783465 100644\r\n--- a/tensorflow/lite/kernels/internal/runtime_shape.h\r\n+++ b/tensorflow/lite/kernels/internal/runtime_shape.h\r\n@@ -44,6 +44,9 @@ class RuntimeShape {\r\n if (dimensions_count > kMaxSmallSize) {\r\n dims_pointer_ = new int32_t[dimensions_count];\r\n }\r\n+ else {\r\n+ dims_ = new int32_t[kMaxSmallSize];\r\n+ }\r\n }\r\n \r\n RuntimeShape(int shape_size, int32_t value) : size_(0) {\r\n@@ -67,6 +70,9 @@ class RuntimeShape {\r\n if (size_ > kMaxSmallSize) {\r\n dims_pointer_ = new int32_t[size_];\r\n }\r\n+ else {\r\n+ dims_ = new int32_t[kMaxSmallSize];\r\n+ }\r\n std::memcpy(DimsData(), other.DimsData(), sizeof(int32_t) * size_);\r\n }\r\n \r\n@@ -80,6 +86,10 @@ class RuntimeShape {\r\n if (size_ > kMaxSmallSize) {\r\n delete[] dims_pointer_;\r\n }\r\n+ else {\r\n+ if (size_ != 0)\r\n+ delete[] dims_;\r\n+ }\r\n }\r\n \r\n inline int32_t DimensionsCount() const { return size_; }\r\n@@ -115,6 +125,9 @@ class RuntimeShape {\r\n if (dimensions_count > kMaxSmallSize) {\r\n dims_pointer_ = new int32_t[dimensions_count];\r\n }\r\n+ else {\r\n+ dims_ = new int32_t[kMaxSmallSize];\r\n+ }\r\n }\r\n \r\n inline void ReplaceWith(int dimensions_count, const int32_t* dims_data) {\r\n@@ -182,7 +195,7 @@ class RuntimeShape {\r\n \r\n int32_t size_;\r\n union {\r\n- int32_t dims_[kMaxSmallSize];\r\n+ int32_t* dims_;\r\n int32_t* dims_pointer_;\r\n };\r\n };\r\n```\r\n\r\n</details>\r\n\r\n4. Comment out `input_shape.Dims(i)` part in `CopyDimsToDesc`.\r\n\r\n<details>\r\n<summary> diff </summary>\r\n\r\n```diff\r\ndiff --git a/tensorflow/lite/kernels/internal/common.h b/tensorflow/lite/kernels/internal/common.h\r\nindex 5e8778f183e..80e4589909d 100644\r\n--- a/tensorflow/lite/kernels/internal/common.h\r\n+++ b/tensorflow/lite/kernels/internal/common.h\r\n@@ -950,7 +950,7 @@ inline void CopyDimsToDesc(const RuntimeShape& input_shape,\r\n NdArrayDesc<N>* desc_out) {\r\n int desc_stride = 1;\r\n for (int i = N - 1; i >= 0; --i) {\r\n- desc_out->extents[i] = input_shape.Dims(i);\r\n+ // desc_out->extents[i] = input_shape.Dims(i);\r\n desc_out->strides[i] = desc_stride;\r\n desc_stride *= input_shape.Dims(i);\r\n }\r\n```\r\n\r\n</details>", "@Seunghui98 Could you please Use a different model quantization method and avoid using models with more than 8 dimensions. Please try with the latest TF version as you are using an older TF version(2.8.0) ? \r\nThank you!", "@sushreebarsa \r\n\r\nThis problem(array out-of-bound) seems to be caused by the gcc version of c++. -> Compiler Bug\r\nIn Ubuntu 18.04, the basic version of gcc is 7.4, and \r\nit seems that there is a bug related to `built_in_memcpy` in tensorflow only in this version. \r\nThis is because the upper versions, Ubuntu 20.04 and 22.04, have a basic version of gcc of 9.4 and no related errors occur here. :)", "@pkgoogle,\r\nPlease look into the issue.\r\n\r\nThank You", "Hi @Seunghui98, is there any reason you can't use a newer version of TF & gcc?\r\n\r\nI should also note that clang will be better supported, now and in the future so it is recommended, especially with versions >= 2.13, to use clang instead of gcc.", "@pkgoogle \r\n\r\n> is there any reason you can't use a newer version of TF & gcc?\r\n\r\nYes. Our project using tflite kernel has dependency on specific `gcc` and `TF` version now.\r\n\r\n> so it is recommended, especially with versions >= 2.13, to use clang instead of gcc.\r\n\r\nOkay.. Thank you for your reply. :)", "Hi @Seunghui98, as you stated yourself gcc == 9.4 already resolves the issue, so I think the solution would be to figure out how to update your project/stack as in some sense this is already fixed right? Let us know if for some reason this does not resolve the issue.\r\n\r\nPlease review the compatibility matrix: https://www.tensorflow.org/install/source#cpu", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62440\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62440\">No</a>\n" ]	2023-11-20T11:47:22	2023-12-16T01:48:29	2023-12-16T01:48:25	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.8.0 ### Custom code Yes ### OS platform and distribution Linux Ubuntu 18.04 ### Mobile device No ### Python version 3.8 ### Bazel version no ### GCC/compiler version gcc 8.4.0 ### CUDA/cuDNN version no ### GPU model and memory no ### Current behavior? - I am using the tensorflow-lite interpreter in the project using [tensorflow/lite/kernels/internal/reference/broadcast_to.h](https://github.com/tensorflow/tensorflow/blob/28b008fdf686a760dd4b4989978bb98d594831ae/tensorflow/lite/kernels/internal/reference/broadcast_to.h#L62). - However, by setting kMaxDims to 8 in file [tensorflow/lite/kernels/broadcast_to.cc](https://github.com/tensorflow/tensorflow/blob/28b008fdf686a760dd4b4989978bb98d594831ae/tensorflow/lite/kernels/broadcast_to.cc#L34), I also want to set kMaxDims to 8 in my project and use `tensorflow/lite/kernels/internal/reference/broadcast_to.h`. - In Ubuntu 20.04 and 22.04, there is no problem using kMaxDims as 8 and calling, but an `out-of-bounds error` occurred in a specific version of `Ubuntu18.04`. - It seems that the size check of the dimension in `tflite::RuntimeShape` exceeds the kMaxSmallSize in [tensorflow/lite/kernels/internal/runtime_shape.h](https://github.com/tensorflow/tensorflow/blob/1b8f5c396f0c016ebe81fe1af029e6f205c926a4/tensorflow/lite/kernels/internal/runtime_shape.h#L37) of 5, but I do not know why this problem occurs only in Ubuntu 18.04. - Could you tell me about this bug? ### Standalone code to reproduce the issue - my codes in samsung ONE project - [link](https://github.com/Samsung/ONE/pull/11701/commits/06164d1a7503e667b5b8206f73fec8a35bbb1b21#diff-df03d71bae09c805951c819c718f8d8eb62fc772a8d3038683c6db1bcba560bb) ```shell /* * Copyright (c) 2023 Samsung Electronics Co., Ltd. All Rights Reserved * * Licensed under the Apache License, Version 2.0 (the "License"); * you may not use this file except in compliance with the License. * You may obtain a copy of the License at * * http://www.apache.org/licenses/LICENSE-2.0 * * Unless required by applicable law or agreed to in writing, software * distributed under the License is distributed on an "AS IS" BASIS, * WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied. * See the License for the specific language governing permissions and * limitations under the License. / #ifndef LUCI_INTERPRETER_PAL_BROADCASTTO_H #define LUCI_INTERPRETER_PAL_BROADCASTTO_H #include <tensorflow/lite/kernels/internal/reference/broadcast_to.h> namespace luci_interpreter_pal { static inline void BroadcastTo(const tflite::RuntimeShape &input_shape, const char input_data, const tflite::RuntimeShape &output_shape, char output_data, TfLiteType data_type) { constexpr int kMaxDims = 8; tflite::reference_ops::BroadcastTo<kMaxDims>(input_shape, input_data, output_shape, output_data, data_type); } } // namespace luci_interpreter_pal #endif // LUCI_INTERPRETER_PAL_BROADCASTTO_H ``` ``` ### Relevant log output ```shell - error logs in Ubuntu 18.04 [2023-11-20T06:36:26.326Z] In member function ‘void luci_interpreter::kernels::BroadcastTo::evalFloat() const’: [2023-11-20T06:36:26.326Z] cc1plus: error: ‘void __builtin_memcpy(void, const void, long unsigned int)’ forming offset [33, 40] is out of the bounds [0, 32] of object ‘<anonymous>’ with type ‘tflite::RuntimeShape’ [-Werror=array-bounds] [2023-11-20T06:36:26.326Z] In file included from /opt/jenkins_agent/workspace/nnfw/master/pr-nncc-release/compiler/luci-interpreter/pal/linux/PALBroadcastTo.h:20, [2023-11-20T06:36:26.326Z] from /opt/jenkins_agent/workspace/nnfw/master/pr-nncc-release/compiler/luci-interpreter/src/kernels/BroadcastTo.cpp:20: [2023-11-20T06:36:26.326Z] /opt/jenkins_agent/workspace/nnfw/master/pr-nncc-release/externals/TENSORFLOW-2.8.0/tensorflow/lite/kernels/internal/reference/broadcast_to.h:68:45: note: ‘<anonymous>’ declared here [2023-11-20T06:36:26.326Z] CopyDimsToDesc(RuntimeShape::ExtendedShape(N, unextended_input_shape), [2023-11-20T06:36:26.326Z] ~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~ [2023-11-20T06:36:26.326Z] cc1plus: error: ‘void* __builtin_memcpy(void, const void, long unsigned int)’ forming offset [33, 40] is out of the bounds [0, 32] of object ‘<anonymous>’ with type ‘tflite::RuntimeShape’ [-Werror=array-bounds] [2023-11-20T06:36:26.326Z] /opt/jenkins_agent/workspace/nnfw/master/pr-nncc-release/externals/TENSORFLOW-2.8.0/tensorflow/lite/kernels/internal/reference/broadcast_to.h:70:45: note: ‘<anonymous>’ declared here [2023-11-20T06:36:26.326Z] CopyDimsToDesc(RuntimeShape::ExtendedShape(N, unextended_output_shape), [2023-11-20T06:36:26.326Z] ~~~~~~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~ [2023-11-20T06:36:26.326Z] [ 15%] Building CXX object compiler/tflchef/tflite/CMakeFiles/tflchef_tflite.dir/src/Op/ReduceMax.cpp.o [2023-11-20T06:36:26.326Z] cc1plus: all warnings being treated as errors [2023-11-20T06:36:26.326Z] compiler/luci-interpreter/src/kernels/CMakeFiles/luci_interpreter_kernels.dir/build.make:278: recipe for target 'compiler/luci-interpreter/src/kernels/CMakeFiles/luci_interpreter_kernels.dir/BroadcastTo.cpp.o' failed [2023-11-20T06:36:26.326Z] make[2]: * [compiler/luci-interpreter/src/kernels/CMakeFiles/luci_interpreter_kernels.dir/BroadcastTo.cpp.o] Error 1 [2023-11-20T06:36:26.326Z] CMakeFiles/Makefile2:8048: recipe for target 'compiler/luci-interpreter/src/kernels/CMakeFiles/luci_interpreter_kernels.dir/all' failed [2023-11-20T06:36:26.326Z] make[1]: * [compiler/luci-interpreter/src/kernels/CMakeFiles/luci_interpreter_kernels.dir/all] Error 2 [2023-11-20T06:36:26.326Z] make[1]: *** Waiting for unfinished jobs.... ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62440/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 1 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62440/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62439	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62439/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62439/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62439/events	https://github.com/tensorflow/tensorflow/issues/62439	2,001,898,905	I_kwDOArmXAs53Uo2Z	62,439	Support TFLite Shared library build with CMake	{ "login": "neg-c", "id": 50472265, "node_id": "MDQ6VXNlcjUwNDcyMjY1", "avatar_url": "https://avatars.githubusercontent.com/u/50472265?v=4", "gravatar_id": "", "url": "https://api.github.com/users/neg-c", "html_url": "https://github.com/neg-c", "followers_url": "https://api.github.com/users/neg-c/followers", "following_url": "https://api.github.com/users/neg-c/following{/other_user}", "gists_url": "https://api.github.com/users/neg-c/gists{/gist_id}", "starred_url": "https://api.github.com/users/neg-c/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/neg-c/subscriptions", "organizations_url": "https://api.github.com/users/neg-c/orgs", "repos_url": "https://api.github.com/users/neg-c/repos", "events_url": "https://api.github.com/users/neg-c/events{/privacy}", "received_events_url": "https://api.github.com/users/neg-c/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473173272, "node_id": "MDU6TGFiZWw0NzMxNzMyNzI=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:feature", "name": "type:feature", "color": "159b2e", "default": false, "description": "Feature requests" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "LakshmiKalaKadali", "id": 149650845, "node_id": "U_kgDOCOt9nQ", "avatar_url": "https://avatars.githubusercontent.com/u/149650845?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LakshmiKalaKadali", "html_url": "https://github.com/LakshmiKalaKadali", "followers_url": "https://api.github.com/users/LakshmiKalaKadali/followers", "following_url": "https://api.github.com/users/LakshmiKalaKadali/following{/other_user}", "gists_url": "https://api.github.com/users/LakshmiKalaKadali/gists{/gist_id}", "starred_url": "https://api.github.com/users/LakshmiKalaKadali/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LakshmiKalaKadali/subscriptions", "organizations_url": "https://api.github.com/users/LakshmiKalaKadali/orgs", "repos_url": "https://api.github.com/users/LakshmiKalaKadali/repos", "events_url": "https://api.github.com/users/LakshmiKalaKadali/events{/privacy}", "received_events_url": "https://api.github.com/users/LakshmiKalaKadali/received_events", "type": "User", "site_admin": false }	[ { "login": "LakshmiKalaKadali", "id": 149650845, "node_id": "U_kgDOCOt9nQ", "avatar_url": "https://avatars.githubusercontent.com/u/149650845?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LakshmiKalaKadali", "html_url": "https://github.com/LakshmiKalaKadali", "followers_url": "https://api.github.com/users/LakshmiKalaKadali/followers", "following_url": "https://api.github.com/users/LakshmiKalaKadali/following{/other_user}", "gists_url": "https://api.github.com/users/LakshmiKalaKadali/gists{/gist_id}", "starred_url": "https://api.github.com/users/LakshmiKalaKadali/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LakshmiKalaKadali/subscriptions", "organizations_url": "https://api.github.com/users/LakshmiKalaKadali/orgs", "repos_url": "https://api.github.com/users/LakshmiKalaKadali/repos", "events_url": "https://api.github.com/users/LakshmiKalaKadali/events{/privacy}", "received_events_url": "https://api.github.com/users/LakshmiKalaKadali/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @neg-c, \r\n\r\nTensorflowlite shared library support is available with Cmake build for C API, and also as this [comment](https://github.com/tensorflow/tensorflow/issues/52542#issuecomment-986341903) suggests please use Bazel for shared library for your usecase.\r\n\r\nThank You\r\n", "Hi @LakshmiKalaKadali thank you, that works.\r\n\r\nI understand that the main build system used for TF is bazel, but having only static build support with CMake doesnt really make sense, or am i missing something? \r\n\r\nPlease let me know if you are planning on adding shared build support, in which case i would be happy to help.", "Hi @neg-c ,\r\n\r\nThanks for your interest in supporting Shared build support for Cmake. Certain complexities are involved in creating c++ shared libraries unlike C where only header files are required. Moreover, compared to Bazel, the way they handle the libraries is different in Cmake. As per my knowledge at that moment, there is no plan in adding shared library support for Cmake. \r\n\r\nThank you.\r\n\r\n", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further." ]	2023-11-20T10:46:11	2023-12-09T01:48:14	2023-12-09T01:48:14	NONE	null	null	null	### Issue type Feature Request ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.14 ### Custom code No ### OS platform and distribution Linux Ubuntu 20.04 ### Mobile device _No response_ ### Python version _No response_ ### Bazel version Cmake 3.27.7 ### GCC/compiler version gcc 9.4.0 ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? By default the Cmake build produces only static library which is not convenient when you have multiple project that use TFLite ### Standalone code to reproduce the issue [TFLite build guide](https://www.tensorflow.org/lite/guide/build_cmake) ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62439/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62439/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62438	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62438/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62438/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62438/events	https://github.com/tensorflow/tensorflow/issues/62438	2,001,476,535	I_kwDOArmXAs53TBu3	62,438	tensorflow - tf.cond does not work if predicate is passed from function	{ "login": "eml-39502", "id": 150397632, "node_id": "U_kgDOCPbiwA", "avatar_url": "https://avatars.githubusercontent.com/u/150397632?v=4", "gravatar_id": "", "url": "https://api.github.com/users/eml-39502", "html_url": "https://github.com/eml-39502", "followers_url": "https://api.github.com/users/eml-39502/followers", "following_url": "https://api.github.com/users/eml-39502/following{/other_user}", "gists_url": "https://api.github.com/users/eml-39502/gists{/gist_id}", "starred_url": "https://api.github.com/users/eml-39502/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/eml-39502/subscriptions", "organizations_url": "https://api.github.com/users/eml-39502/orgs", "repos_url": "https://api.github.com/users/eml-39502/repos", "events_url": "https://api.github.com/users/eml-39502/events{/privacy}", "received_events_url": "https://api.github.com/users/eml-39502/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1097545817, "node_id": "MDU6TGFiZWwxMDk3NTQ1ODE3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:apis", "name": "comp:apis", "color": "0052cc", "default": false, "description": "Highlevel API related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false }	[ { "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false } ]	null	[ "@eml-39502,\r\nLooks like this is a duplicate of the issue [#57492](https://github.com/tensorflow/tensorflow/issues/57492). Could you please check this issue, and confirm the same? Thank you!\r\n", "Hi, \r\nFirst of all, I must warn you that I am not an expert on the subject, so I apologize in advance if my answer lacks any coherence.\r\n\r\nIn my opinion the problems with your code are the following:\r\n\r\n1. apparently the tf.cond() function executes both internal functions (true_fn and false_fn), somehow mapping the results and selecting them after checking the \"pred\" or flag.\r\n2. flat_map() tries to return a dataset but one of the routes returns nothing (false_fn), and even worst, this function cuts the execution with an assert. Thus generating a conflict\r\n\r\nThe behavior of your code is then to fail when executing the false_fn branch and after that executes the true_fn since the predicate is a constant with value true.\r\n\r\n\r\nHere my solutions:\r\n\r\nBy using tf.cond\r\n``` python\r\nimport tensorflow as tf\r\nimport tensorflow_datasets as tfds\r\n\r\ndef do_not_come_here(image, flag):\r\n tf.print(flag)\r\n return image\r\n\r\ndef g(image, flag):\r\n return tf.cond(flag, lambda: image, lambda: do_not_come_here(image, flag))\r\n\r\ndef f(record):\r\n image = record['image']\r\n flag = tf.constant(True, dtype=tf.bool)\r\n processed_image = g(image, flag)\r\n processed_dataset = tf.data.Dataset.from_tensor_slices({\r\n 'processed_image': processed_image\r\n })\r\n return processed_dataset\r\n\r\ntrain, info = tfds.load(\"voc\", split='train', with_info=True)\r\ndataset = train.take(1).flat_map(f)\r\n\r\n#To display the tf.print()\r\nfor example in dataset:\r\n pass\r\n```\r\n\r\nBy using tf.py_function\r\n``` python\r\nimport tensorflow as tf\r\nimport tensorflow_datasets as tfds\r\n\r\ndef do_not_come_here(image, flag):\r\n tf.print(flag)\r\n tf.debugging.assert_equal(0, 1)\r\n return image\r\n\r\ndef g(image, flag):\r\n return tf.py_function(\r\n func=lambda img, flg: img if flg else do_not_come_here(img, flg),\r\n inp=[image, flag],\r\n Tout=tf.uint8\r\n )\r\n\r\ndef f(record):\r\n image = record['image']\r\n flag = tf.constant(True, dtype=tf.bool)\r\n processed_image = g(image, flag)\r\n processed_dataset = tf.data.Dataset.from_tensor_slices({\r\n 'processed_image': processed_image\r\n })\r\n return processed_dataset\r\n\r\ntrain, info = tfds.load(\"voc\", split='train', with_info=True)\r\ndataset = train.take(1).flat_map(f)\r\n```", "Hi @tilakrayal \r\n> @eml-39502, Looks like this is a duplicate of the issue [#57492](https://github.com/tensorflow/tensorflow/issues/57492). Could you please check this issue, and confirm the same? Thank you!\r\n\r\nI think the issue 57492 is having ```InvalidArgumentError: Condition x > y did not hold element-wise:``` exception. In my issue, I am casing the exception to confirm TF execution came to the place it should not have come to.\r\n\r\nThe issue here is ```tf.cond``` executes both ```true_fn``` and ```false_fn``` regardless with the predicate, hence it is not working as conditional branching function.\r\n\r\nHope this clarifies.\r\n", "@eml-39502,\r\nHave you got the chance to have a look at the provided alternative workaround where the mentioned code was executed without any issue/error. Kindly find the gist of it [here](https://colab.research.google.com/gist/tilakrayal/c0261efbbd7cafd4ca48e703d03e7312/untitled1556.ipynb) and also the tf.cond() function evaluates both the `true_fn` and `false_fn`, regardless of the value of the pred argument. The resulting tensors are then mapped to the appropriate output tensor based on the value of pred. Thank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62438\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62438\">No</a>\n" ]	2023-11-20T06:39:55	2023-12-21T01:48:43	2023-12-21T01:48:39	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source binary ### TensorFlow version 2.14.1 ### Custom code No ### OS platform and distribution NAME="Pop!_OS" VERSION="22.04 LTS" (Ubuntu equivalent) ### Mobile device _No response_ ### Python version 3.10.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version 11.8 ### GPU model and memory _No response_ ### Current behavior? The code below somehow always goes to ```do_not_come_here```. I suppose the usage ```tf.cond``` in the function ```g``` is not correct but not sure how to fix. Please advise. ## Code ## ``` import tensorflow as tf import tensorflow_datasets as tfds def do_not_come_here(image, flag): tf.print(flag) tf.assert_equal(0, 1) def g(image, flag): tf.cond( pred=flag, # same with pred=tf.math.equal(flag, tf.constant(True)) true_fn=lambda: image, false_fn=lambda: do_not_come_here(image, flag) ) def f(record): image = record['image'] g(image, tf.constant(True)) train, info = tfds.load("voc", split='train', with_info=True) train.take(1).flat_map(f) # same with .map(f) ``` ## Stacktrace ``` ~/venv/ml/bin/python ~/reproduction.py 2023-11-20 17:19:25.390562: I tensorflow/core/util/port.cc:111] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. 2023-11-20 17:19:25.415972: E tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:9342] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered 2023-11-20 17:19:25.415993: E tensorflow/compiler/xla/stream_executor/cuda/cuda_fft.cc:609] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered 2023-11-20 17:19:25.416010: E tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:1518] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered 2023-11-20 17:19:25.420375: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. 2023-11-20 17:19:26.669052: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.672374: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.672502: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.673349: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.673435: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.673493: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.729836: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.729955: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.730043: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.730112: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1886] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 1228 MB memory: -> device: 0, name: NVIDIA GeForce RTX 4050 Laptop GPU, pci bus id: 0000:01:00.0, compute capability: 8.9 Traceback (most recent call last): File "~/reproduction.py", line 21, in <module> train.take(1).flat_map(f) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/dataset_ops.py", line 2313, in flat_map return flat_map_op._flat_map(self, map_func, name=name) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/flat_map_op.py", line 24, in _flat_map return _FlatMapDataset(input_dataset, map_func, name) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/flat_map_op.py", line 33, in __init__ self._map_func = structured_function.StructuredFunctionWrapper( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/structured_function.py", line 265, in __init__ self._function = fn_factory() File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py", line 1222, in get_concrete_function concrete = self._get_concrete_function_garbage_collected(args, kwargs) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py", line 1192, in _get_concrete_function_garbage_collected self._initialize(args, kwargs, add_initializers_to=initializers) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py", line 694, in _initialize self._concrete_variable_creation_fn = tracing_compilation.trace_function( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/tracing_compilation.py", line 178, in trace_function concrete_function = _maybe_define_function( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/tracing_compilation.py", line 284, in _maybe_define_function concrete_function = _create_concrete_function( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/tracing_compilation.py", line 308, in _create_concrete_function traced_func_graph = func_graph_module.func_graph_from_py_func( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/framework/func_graph.py", line 1059, in func_graph_from_py_func func_outputs = python_func(func_args, *func_kwargs) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py", line 597, in wrapped_fn out = weak_wrapped_fn().__wrapped__(args, *kwds) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/structured_function.py", line 231, in wrapped_fn ret = wrapper_helper(args) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/structured_function.py", line 161, in wrapper_helper ret = autograph.tf_convert(self._func, ag_ctx)(nested_args) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/autograph/impl/api.py", line 693, in wrapper raise e.ag_error_metadata.to_exception(e) tensorflow.python.framework.errors_impl.InvalidArgumentError: in user code: File "~/reproduction.py", line 18, in f g(image, tf.constant(True)) File "~/reproduction.py", line 10, in g * tf.cond( File "~/reproduction.py", line 7, in do_not_come_here * tf.assert_equal(0, 1) InvalidArgumentError: Condition x == y did not hold element-wise: x (cond/assert_equal_1/x:0) = 0 y (cond/assert_equal_1/y:0) = 1 Process finished with exit code 1 ``` Same symptom with: ``` def g(image, flag): tf.cond( pred=tf.math.equal(flag, tf.constant(1)), true_fn=lambda: image, false_fn=lambda: do_not_come_here(image, flag) ) def f(record): image = record['image'] g(image, tf.constant(1)) ``` Same symptom with: ``` def g(image, flag): tf.cond( pred=tf.math.equal(flag, tf.constant(1)), true_fn=lambda: do_not_come_here(image, flag), # <--- swap true_fn with false_fn false_fn=lambda: image ) def f(record): image = record['image'] g(image, tf.constant(1)) ``` ## Environment ``` Python 3.10.12 tensorflow 2.14.1 tensorflow-datasets 4.9.3 NAME="Pop!_OS" VERSION="22.04 LTS" ``` ### Standalone code to reproduce the issue ```shell import tensorflow as tf import tensorflow_datasets as tfds def do_not_come_here(image, flag): tf.print(flag) tf.assert_equal(0, 1) def g(image, flag): tf.cond( pred=flag, # same with pred=tf.math.equal(flag, tf.constant(True)) true_fn=lambda: image, false_fn=lambda: do_not_come_here(image, flag) ) def f(record): image = record['image'] g(image, tf.constant(True)) train, info = tfds.load("voc", split='train', with_info=True) train.take(1).flat_map(f) # same with .map(f) ``` ### Relevant log output ```shell ~/venv/ml/bin/python ~/reproduction.py 2023-11-20 17:19:25.390562: I tensorflow/core/util/port.cc:111] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`. 2023-11-20 17:19:25.415972: E tensorflow/compiler/xla/stream_executor/cuda/cuda_dnn.cc:9342] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered 2023-11-20 17:19:25.415993: E tensorflow/compiler/xla/stream_executor/cuda/cuda_fft.cc:609] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered 2023-11-20 17:19:25.416010: E tensorflow/compiler/xla/stream_executor/cuda/cuda_blas.cc:1518] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered 2023-11-20 17:19:25.420375: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. 2023-11-20 17:19:26.669052: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.672374: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.672502: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.673349: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.673435: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.673493: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.729836: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.729955: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.730043: I tensorflow/compiler/xla/stream_executor/cuda/cuda_gpu_executor.cc:894] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355 2023-11-20 17:19:26.730112: I tensorflow/core/common_runtime/gpu/gpu_device.cc:1886] Created device /job:localhost/replica:0/task:0/device:GPU:0 with 1228 MB memory: -> device: 0, name: NVIDIA GeForce RTX 4050 Laptop GPU, pci bus id: 0000:01:00.0, compute capability: 8.9 Traceback (most recent call last): File "~/reproduction.py", line 21, in <module> train.take(1).flat_map(f) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/dataset_ops.py", line 2313, in flat_map return flat_map_op._flat_map(self, map_func, name=name) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/flat_map_op.py", line 24, in _flat_map return _FlatMapDataset(input_dataset, map_func, name) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/flat_map_op.py", line 33, in __init__ self._map_func = structured_function.StructuredFunctionWrapper( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/structured_function.py", line 265, in __init__ self._function = fn_factory() File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py", line 1222, in get_concrete_function concrete = self._get_concrete_function_garbage_collected(args, kwargs) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py", line 1192, in _get_concrete_function_garbage_collected self._initialize(args, kwargs, add_initializers_to=initializers) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py", line 694, in _initialize self._concrete_variable_creation_fn = tracing_compilation.trace_function( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/tracing_compilation.py", line 178, in trace_function concrete_function = _maybe_define_function( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/tracing_compilation.py", line 284, in _maybe_define_function concrete_function = _create_concrete_function( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/tracing_compilation.py", line 308, in _create_concrete_function traced_func_graph = func_graph_module.func_graph_from_py_func( File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/framework/func_graph.py", line 1059, in func_graph_from_py_func func_outputs = python_func(func_args, *func_kwargs) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/eager/polymorphic_function/polymorphic_function.py", line 597, in wrapped_fn out = weak_wrapped_fn().__wrapped__(args, *kwds) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/structured_function.py", line 231, in wrapped_fn ret = wrapper_helper(args) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/data/ops/structured_function.py", line 161, in wrapper_helper ret = autograph.tf_convert(self._func, ag_ctx)(nested_args) File "~/venv/ml/lib/python3.10/site-packages/tensorflow/python/autograph/impl/api.py", line 693, in wrapper raise e.ag_error_metadata.to_exception(e) tensorflow.python.framework.errors_impl.InvalidArgumentError: in user code: File "~/reproduction.py", line 18, in f g(image, tf.constant(True)) File "~/reproduction.py", line 10, in g * tf.cond( File "~/reproduction.py", line 7, in do_not_come_here * tf.assert_equal(0, 1) InvalidArgumentError: Condition x == y did not hold element-wise: x (cond/assert_equal_1/x:0) = 0 y (cond/assert_equal_1/y:0) = 1 Process finished with exit code 1 ``` ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62438/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62438/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62437	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62437/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62437/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62437/events	https://github.com/tensorflow/tensorflow/pull/62437	2,001,362,088	PR_kwDOArmXAs5f3aqy	62,437	Remove endian conversion for S4/U4 literals for s390x	{ "login": "yasiribmcon", "id": 107119183, "node_id": "U_kgDOBmKCTw", "avatar_url": "https://avatars.githubusercontent.com/u/107119183?v=4", "gravatar_id": "", "url": "https://api.github.com/users/yasiribmcon", "html_url": "https://github.com/yasiribmcon", "followers_url": "https://api.github.com/users/yasiribmcon/followers", "following_url": "https://api.github.com/users/yasiribmcon/following{/other_user}", "gists_url": "https://api.github.com/users/yasiribmcon/gists{/gist_id}", "starred_url": "https://api.github.com/users/yasiribmcon/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/yasiribmcon/subscriptions", "organizations_url": "https://api.github.com/users/yasiribmcon/orgs", "repos_url": "https://api.github.com/users/yasiribmcon/repos", "events_url": "https://api.github.com/users/yasiribmcon/events{/privacy}", "received_events_url": "https://api.github.com/users/yasiribmcon/received_events", "type": "User", "site_admin": false }	[ { "id": 1133285679, "node_id": "MDU6TGFiZWwxMTMzMjg1Njc5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:xla", "name": "comp:xla", "color": "0052cc", "default": false, "description": "XLA" }, { "id": 1169364458, "node_id": "MDU6TGFiZWwxMTY5MzY0NDU4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:S", "name": "size:S", "color": "adafea", "default": false, "description": "CL Change Size: Small" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @cheshire ,\r\nCould you please kindly review this PR? Thank you very much!", "@reedwm WDYT?", "I think XLA changes must be submitted to the [openxla/xla](https://github.com/openxla/xla) repo, not the TensorFlow repo (CC @ddunl, correct me if I'm wrong). @yasiribmcon, can you create a new PR to the openxla/xla repo with this change and CC me? Once merged, the TF repo will automatically be updated.", "Yeah in principle it should be possible from here, but we have it disabled for at least the next month. So it would be great if you could reopen on openxla/xla. Feel free to tag me and I'll make sure it gets merged quickly.", "Hi @yasiribmcon Any update on this PR? Please. Thank you!", "Apologies for the delay due to christmas break @gbaned @reedwm @ddunl.\r\nCreated PR https://github.com/openxla/xla/pull/8162.", "Hi @gbaned Required changes have been merged with https://github.com/openxla/xla/pull/8162.\r\nPlease feel free to close this PR.\r\nThank you very much!", "> Hi @gbaned Required changes have been merged with [openxla/xla#8162](https://github.com/openxla/xla/pull/8162). Please feel free to close this PR. Thank you very much!\r\n\r\nHi @yasiribmcon Thank you very much for the confirmation. " ]	2023-11-20T04:51:20	2024-01-10T05:08:45	2024-01-10T05:06:41	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62437", "html_url": "https://github.com/tensorflow/tensorflow/pull/62437", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62437.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62437.patch", "merged_at": null }	After introducing [_commit_](https://github.com/linux-on-ibm-z/tensorflow/commit/e0059bba8cb4d6d3979458c0945c66b4cc1c9f69) to add support for int4(S4 and U4) types in literal, `//tensorflow/compiler/xla:literal_test` started to fail for BE(s390x) machines. This issue is occurring in `ProtoRoundTrip` test case when we try to convert/retrieve S4/U4 literal to/from proto values. > auto vector_s4 = LiteralUtil::CreateR1<s4>({s4{-1}, s4{3}, s4{7}}); > auto vector_u4 = LiteralUtil::CreateR1<u4>({u4{1}, u4{3}, u4{15}}); > EXPECT_EQ(vector_s4, to_from_proto(vector_s4)); > EXPECT_EQ(vector_u4, to_from_proto(vector_u4)); ConvertEndianShort assertion is failing as it expects an even byte size, but for vector_s4 / vector_u4, byte size is coming as 3(odd), causing the test case to fail. ConvertEndianShort is getting used mostly to convert 16 bit literal types to LE byte order as protobuf seems to be processing bytes in LE format. > In third_party/xla/xla/xla_data.proto - > message LiteralProto { > ... > bytes s4s = 21; > bytes u4s = 22; > ... > // The F16s, BF16s, U16s and S16s are encoded in little endian byte order > bytes f16s = 11; > bytes bf16s = 13; > bytes u16s = 16; > bytes s16s = 17; Since S4/U4 underlying type is a [single](https://pypi.org/project/ml-dtypes/#:~:text=exponent%20is%200-,int4%20and%20uint4,-4%2Dbit%20integer) byte, endian conversion is not needed for S4/U4 literal types. Hence removing endian conversion for S4/U4 integer literals types. > using u4 = ml_dtypes::uint4; > using s4 = ml_dtypes::int4; After these changes, `//tensorflow/compiler/xla:literal_test` test case passes. These changes do not cause any regressions on existing test cases and it won't affect LE machines.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62437/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62437/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62436	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62436/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62436/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62436/events	https://github.com/tensorflow/tensorflow/issues/62436	2,001,248,351	I_kwDOArmXAs53SKBf	62,436	Bug in tf.test.compute_gradient for tf.keras.layers.LeakyReLU	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @cheyennee ,\r\n\r\nI have replicated the reported error and attached the gist [here](https://colab.sandbox.google.com/gist/SuryanarayanaY/adf56731c80d34c22f6bc9cb774bae76/62436.ipynb).Need to check for the difference in results." ]	2023-11-20T02:34:24	2023-11-22T07:31:05	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? The numerical gradient and theoretical gradient for `tf.keras.layers.LeakyReLU` is different, one is `0.2`, the other is `0.6`. ### Standalone code to reproduce the issue ```shell import tensorflow as tf alpha = 0.2 __input___0_tensor = tf.random.uniform([2, 1, 1, 1], minval=0, maxval=0, dtype=tf.float64) __input___0 = tf.identity( __input___0_tensor) LeakyReLU_class = tf.keras.layers.LeakyReLU(alpha=alpha, dtype=tf.float64) layer = LeakyReLU_class inputs = __input___0 r = LeakyReLU_class(inputs) theoretical, numerical = tf.test.compute_gradient(LeakyReLU_class, [inputs]) print(theoretical) print(numerical) ``` ### Relevant log output ```shell (array([[0.2, 0. ], [0. , 0.2]]),) (array([[0.6, 0. ], [0. , 0.6]]),) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62436/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62436/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62435	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62435/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62435/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62435/events	https://github.com/tensorflow/tensorflow/issues/62435	2,000,840,271	I_kwDOArmXAs53QmZP	62,435	Bug in tf.test.compute_gradient for tf.keras.layers.GlobalMaxPooling2D	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Same problem can be found in tf.keras.layers.GlobalMaxPooling3D. Here is the repo code:\r\n```\r\nimport tensorflow as tf\r\n\r\ndata_format = \"channels_first\"\r\nkeepdims = False\r\n__input___0_tensor = tf.random.uniform([2, 4, 3, 1, 1], minval=0, maxval=0, dtype=tf.float64)\r\n__input___0 = tf.identity(__input___0_tensor)\r\nGlobalMaxPooling3D_class = tf.keras.layers.GlobalMaxPooling3D(data_format=data_format, keepdims=keepdims)\r\n\r\nlayer = GlobalMaxPooling3D_class\r\ninputs = __input___0\r\n\r\nr = GlobalMaxPooling3D_class(inputs)\r\ntheoretical, numerical = tf.test.compute_gradient(GlobalMaxPooling3D_class, [inputs])\r\nprint(theoretical)\r\nprint(numerical)\r\n```", "Same problem can be found in tf.keras.layers.MaxPooling1D. Here is the repo code:\r\n```\r\npool_size_0 = 2\r\npool_size = [pool_size_0, ]\r\nstrides_0 = 3\r\nstrides = [\r\n strides_0, ]\r\npadding = \"valid\"\r\ndata_format = \"channels_last\"\r\n__input___0_tensor = tf.random.uniform([3, 5, 4], minval=0.0, maxval=0.0, dtype=tf.float64)\r\n__input___0 = tf.identity(__input___0_tensor)\r\nMaxPooling1D_class = tf.keras.layers.MaxPooling1D(pool_size=pool_size, strides=strides, padding=padding, data_format=data_format)\r\n\r\nlayer = MaxPooling1D_class\r\ninputs = __input___0\r\n\r\nr = MaxPooling1D_class(inputs)\r\ntheoretical, numerical = tf.test.compute_gradient(MaxPooling1D_class, [inputs])\r\nprint(theoretical)\r\nprint(numerical)\r\n```", "@SuryanarayanaY I was able to replicate the issue on colab, please find the [gist](https://colab.research.google.com/gist/sushreebarsa/28c3f8073fe8fcd7c9d62212fdec1e0f/62435.ipynb) here.\r\nThank you!" ]	2023-11-19T12:58:32	2023-11-22T07:32:35	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? The numerical gradient and theoretical gradient for `tf.keras.layers.GlobalMaxPooling2D` is different, one is `0.2`, the other is `0.5`. ### Standalone code to reproduce the issue ```shell import tensorflow as tf data_format = "channels_last" keepdims = False __input___0_tensor = tf.random.uniform([3, 5, 1, 1], minval=0, maxval=0, dtype=tf.float64) __input___0 = tf.identity(__input___0_tensor) GlobalMaxPooling2D_class = tf.keras.layers.GlobalMaxPooling2D(data_format=data_format, keepdims=keepdims) layer = GlobalMaxPooling2D_class inputs = __input___0 r = GlobalMaxPooling2D_class(inputs) theoretical, numerical = tf.test.compute_gradient(GlobalMaxPooling2D_class, [inputs]) print(theoretical) print(numerical) ``` ### Relevant log output ```shell (array([[0.2, 0.2, 0.2, 0.2, 0.2, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0.2, 0.2, 0.2, 0.2, 0.2, 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.2, 0.2, 0.2, 0.2, 0.2]]),) (array([[0.5, 0.5, 0.5, 0.5, 0.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0.5, 0.5, 0.5, 0.5, 0.5, 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.5, 0.5, 0.5, 0.5, 0.5]]),) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62435/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62435/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62434	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62434/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62434/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62434/events	https://github.com/tensorflow/tensorflow/issues/62434	2,000,828,824	I_kwDOArmXAs53QjmY	62,434	Bug in tf.test.compute_gradient for tf.keras.layers.ELU	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @cheyennee ,\r\n\r\nI have replicated the reported behaviour with colab using TF v2.14, 2.15, and TF-nightly. Please find the [gist](https://colab.research.google.com/gist/Venkat6871/eb8aed65fbb11f4809db3611979e22ae/62434_2-14_2-15-nightly-v.ipynb) here for reference.\r\n\r\nThank you!", "Thank you for your report!\r\n\r\nThe gradient of ELU is `1 if z > 0 else alpha*exp(z)`. Note that it discontinuous at 0 when alpha != 1.\r\n\r\nHere's an example showing behavior around zero:\r\n\r\n```\r\nimport tensorflow as tf\r\nalpha = 14.5\r\nfor v in [-0.001, 0.0, 0.001]:\r\n inputs = tf.constant(v, shape=[2], dtype=tf.float32)\r\n ELU_class = tf.keras.layers.ELU(alpha=alpha, dtype=tf.float64)\r\n\r\n theoretical, numerical = tf.test.compute_gradient(ELU_class, [inputs])\r\n print(\"\\ninput value:\", v)\r\n print(theoretical)\r\n print(numerical)\r\n```\r\n\r\n```\r\ninput value: -0.001\r\n(array([[14.485507, 0. ],\r\n [ 0. , 14.485507]], dtype=float32),)\r\n(array([[14.485511, 0. ],\r\n [ 0. , 14.485511]], dtype=float32),)\r\n\r\ninput value: 0.0\r\n(array([[14.5, 0. ],\r\n [ 0. , 14.5]], dtype=float32),)\r\n(array([[7.746461, 0. ],\r\n [0. , 7.746461]], dtype=float32),)\r\n\r\ninput value: 0.001\r\n(array([[1., 0.],\r\n [0., 1.]], dtype=float32),)\r\n(array([[1., 0.],\r\n [0., 1.]], dtype=float32),)\r\n```\r\n\r\nThe theoretical value at 0 is alpha = 14.5 as expected. The numeric value at 0 is the average of the values on both sides, (1 + alpha) / 2 = (1 + 14.5) / 2 = 7.75 as expected." ]	2023-11-19T12:24:23	2023-12-15T18:09:11	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? The numerical gradient and theoretical gradient for `tf.keras.layers.ELU` is different, one is `14.5`, the other is `7.746461`. ### Standalone code to reproduce the issue ```shell import tensorflow as tf alpha = 14.5 __input___0_tensor = tf.random.uniform([2, 3, 4], minval=0, maxval=0, dtype=tf.float32) __input___0 = tf.identity( __input___0_tensor) ELU_class = tf.keras.layers.ELU(alpha=alpha, dtype=tf.float64) layer = ELU_class inputs = __input___0 r = ELU_class(inputs) theoretical, numerical = tf.test.compute_gradient(ELU_class, [inputs]) print(theoretical) print(numerical) ``` ### Relevant log output ```shell (array([[14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. , 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5, 0. ], [ 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 14.5]], dtype=float32),) (array([[7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461, 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 7.746461]], dtype=float32),) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62434/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62434/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62433	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62433/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62433/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62433/events	https://github.com/tensorflow/tensorflow/issues/62433	2,000,820,517	I_kwDOArmXAs53Qhkl	62,433	tf.keras.layers.DepthwiseConv2D throws error when backprop	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@cheyennee,\r\nI was able to replicate the reported behaviour with TF v2.13, TF v2.14 and tf-nightly. Kindly find the [gist](https://colab.research.google.com/gist/tilakrayal/db2eebf2422a7fb876d84736c20b875c/untitled1523.ipynb) for reference.\r\nCurrently we are investigating the issue & will deep dive into the issue and provide the root-cause for the same. Thank you!\r\n" ]	2023-11-19T12:00:24	2023-12-07T22:13:37	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? tf.keras.layers.DepthwiseConv2D throws error in backprop. ### Standalone code to reproduce the issue ```shell kernel_size_0 = 3 kernel_size_1 = 3 kernel_size = [kernel_size_0,kernel_size_1, ] strides_0 = 2 strides_1 = 2 strides = [ strides_0, strides_1, ] padding = "valid" depth_multiplier = 1 data_format = None dilation_rate_0 = 1 dilation_rate_1 = 1 dilation_rate = [dilation_rate_0,dilation_rate_1,] activation = None use_bias = False depthwise_initializer = "glorot_uniform" bias_initializer = "zeros" depthwise_regularizer = None bias_regularizer = None activity_regularizer = None depthwise_constraint = None bias_constraint = None __input___0_tensor = tf.random.uniform([2, 15, 1, 1], minval=0, maxval=0, dtype=tf.float64) __input___0 = tf.identity(__input___0_tensor) DepthwiseConv2D_class = tf.keras.layers.DepthwiseConv2D(kernel_size, strides=strides, padding=padding, depth_multiplier=depth_multiplier, data_format=data_format, dilation_rate=dilation_rate, activation=activation, use_bias=use_bias, depthwise_initializer=depthwise_initializer, bias_initializer=bias_initializer, depthwise_regularizer=depthwise_regularizer, bias_regularizer=bias_regularizer, activity_regularizer=activity_regularizer, depthwise_constraint=depthwise_constraint, bias_constraint=bias_constraint) layer = DepthwiseConv2D_class inputs = __input___0 with tf.GradientTape() as g: g.watch(inputs) res = layer(inputs) grad = g.jacobian(res, inputs) ``` ### Relevant log output ```shell --------------------------------------------------------------------------- InvalidArgumentError Traceback (most recent call last) <ipython-input-3-def5aab96c99> in <cell line: 34>() 32 g.watch(inputs) 33 res = layer(inputs) ---> 34 grad = g.jacobian(res, inputs) 3 frames /usr/local/lib/python3.10/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name) 58 for t in inputs 59 ] ---> 60 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name, 61 inputs, attrs, num_outputs) 62 except core._NotOkStatusException as e: InvalidArgumentError: Graph execution error: Detected at node gradient_tape/DepthwiseConv2dNativeBackpropInput/pfor/TensorArrayV2Stack/TensorListStack defined at (most recent call last): File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main File "/usr/lib/python3.10/runpy.py", line 86, in _run_code File "/usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py", line 37, in <module> File "/usr/local/lib/python3.10/dist-packages/traitlets/config/application.py", line 992, in launch_instance File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelapp.py", line 619, in start File "/usr/local/lib/python3.10/dist-packages/tornado/platform/asyncio.py", line 195, in start File "/usr/lib/python3.10/asyncio/base_events.py", line 603, in run_forever File "/usr/lib/python3.10/asyncio/base_events.py", line 1909, in _run_once File "/usr/lib/python3.10/asyncio/events.py", line 80, in _run File "/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py", line 685, in <lambda> File "/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py", line 738, in _run_callback File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 825, in inner File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 786, in run File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 361, in process_one File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 261, in dispatch_shell File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 539, in execute_request File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py", line 302, in do_execute File "/usr/local/lib/python3.10/dist-packages/ipykernel/zmqshell.py", line 539, in run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 2975, in run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3030, in _run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/async_helpers.py", line 78, in _pseudo_sync_runner File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3257, in run_cell_async File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3473, in run_ast_nodes File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3553, in run_code File "<ipython-input-3-def5aab96c99>", line 34, in <cell line: 34> File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/ops/parallel_for/control_flow_ops.py", line 212, in f Tried to stack elements of an empty list with non-fully-defined element_shape: [?,?,?,?] [[{{node gradient_tape/DepthwiseConv2dNativeBackpropInput/pfor/TensorArrayV2Stack/TensorListStack}}]] [Op:__inference_f_345] ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62433/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62433/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62432	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62432/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62432/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62432/events	https://github.com/tensorflow/tensorflow/issues/62432	2,000,814,982	I_kwDOArmXAs53QgOG	62,432	tf.keras.layers.Dense throws error when backprop	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @cheyennee ,\r\n\r\nI have replicated the reported behaviour with TF2.14 and tf-nightly as well. Attached [gist](https://colab.sandbox.google.com/gist/SuryanarayanaY/5df36464b3a5a49174b297d517224e81/62432.ipynb) for reference.\r\n\r\nWill dig more into for the root cause and come back. Thanks!", "Hi @SuryanarayanaY ,\r\nI believe the bug stems from the absence of units validation. As per the [documentation](https://www.tensorflow.org/api_docs/python/tf/keras/layers/Dense), units should be a positive integer. However, in the provided code, when units is set to 0, no error is raised during the execution of res = layer(inputs).", "@cheyennee ,\r\n\r\nMaking `units>0` indeed resolves this issue. I can see validation for `units < 0` not `units <= 0`. This needs to check internally. " ]	2023-11-19T11:44:12	2024-01-11T09:49:12	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? tf.keras.layers.Dense crashes in backprop. ### Standalone code to reproduce the issue ```shell import tensorflow as tf units = 0 activation = "sigmoid" use_bias = True kernel_initializer = "ones" bias_initializer = "zeros" kernel_regularizer = None bias_regularizer = None activity_regularizer = None kernel_constraint = None bias_constraint = None __input___0_tensor = tf.random.uniform([50, 1], minval=1.0, maxval=3.0, dtype=tf.float64) __input___0 = tf.identity(__input___0_tensor) Dense_class = tf.keras.layers.Dense(units, activation=activation, use_bias=use_bias, kernel_initializer=kernel_initializer, bias_initializer=bias_initializer, kernel_regularizer=kernel_regularizer, bias_regularizer=bias_regularizer, activity_regularizer=activity_regularizer, kernel_constraint=kernel_constraint, bias_constraint=bias_constraint) layer = Dense_class inputs = __input___0 with tf.GradientTape() as g: g.watch(inputs) res = layer(inputs) print(res.shape) grad = g.jacobian(res, inputs) # Error print(grad) ``` ### Relevant log output ```shell (50, 0) --------------------------------------------------------------------------- ValueError Traceback (most recent call last) <ipython-input-2-b51b65c26b37> in <cell line: 23>() 21 res = layer(inputs) 22 print(res.shape) ---> 23 grad = g.jacobian(res, inputs) # Error 24 print(grad) 1 frames /usr/local/lib/python3.10/dist-packages/tensorflow/python/framework/ops.py in set_shape(self, shape) 534 def set_shape(self, shape) -> None: 535 if not self.shape.is_compatible_with(shape): --> 536 raise ValueError(f"Tensor's shape {self.shape} is not compatible " 537 f"with supplied shape {shape}.") 538 ValueError: Tensor's shape (50, 0, 50, 0) is not compatible with supplied shape (50, 0, 50, 1). ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62432/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62432/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62431	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62431/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62431/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62431/events	https://github.com/tensorflow/tensorflow/issues/62431	2,000,772,143	I_kwDOArmXAs53QVwv	62,431	Bug in tf.test.compute_gradient for tf.keras.layers.Conv3DTranspose	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@sachinprasadhs I was able to replicate this issue, please find the attached [gist](https://colab.research.google.com/gist/sushreebarsa/05ec89ce86ecb20f3078f4a704a642c7/62431.ipynb) here. \r\nThank you!" ]	2023-11-19T09:33:39	2023-12-26T22:49:18	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? The numerical gradient and theoretical gradient for `tf.keras.layers.Conv3DTranspose` is different. ### Standalone code to reproduce the issue ```shell filters = 2 kernel_size_0 = 3 kernel_size_1 = 3 kernel_size_2 = 3 kernel_size = [kernel_size_0, kernel_size_1,kernel_size_2, ] strides_0 = 2 strides_1 = 2 strides_2 = 2 strides = [ strides_0, strides_1, strides_2, ] padding = "valid" output_padding = None data_format = "channels_last" dilation_rate_0 = 1 dilation_rate_1 = 1 dilation_rate_2 = 1 dilation_rate = [dilation_rate_0,dilation_rate_1,dilation_rate_2,] activation = "relu" use_bias = False kernel_initializer = None bias_initializer = None kernel_regularizer = None bias_regularizer = None activity_regularizer = None kernel_constraint = None bias_constraint = None __input___0_tensor = tf.random.uniform([3, 5, 1, 1, 1], minval=0, maxval=0, dtype=tf.float64) __input___0 = tf.identity(__input___0_tensor) Conv3DTranspose_class = tf.keras.layers.Conv3DTranspose(filters, kernel_size, strides=strides, padding=padding, output_padding=output_padding, data_format=data_format, dilation_rate=dilation_rate, activation=activation, use_bias=use_bias, kernel_initializer=kernel_initializer, bias_initializer=bias_initializer, kernel_regularizer=kernel_regularizer, bias_regularizer=bias_regularizer, activity_regularizer=activity_regularizer, kernel_constraint=kernel_constraint, bias_constraint=bias_constraint) layer = Conv3DTranspose_class inputs = __input___0 r = Conv3DTranspose_class(inputs) theoretical, numerical = tf.test.compute_gradient(Conv3DTranspose_class, [inputs]) print(theoretical) print(numerical) ``` ### Relevant log output ```shell (array([[0., 0., 0., ..., 0., 0., 0.], [0., 0., 0., ..., 0., 0., 0.], [0., 0., 0., ..., 0., 0., 0.], ..., [0., 0., 0., ..., 0., 0., 0.], [0., 0., 0., ..., 0., 0., 0.], [0., 0., 0., ..., 0., 0., 0.]]),) (array([[-0.05650689, 0. , 0. , ..., 0. , 0. , 0. ], [-0.11539936, 0. , 0. , ..., 0. , 0. , 0. ], [ 0.06827629, 0. , 0. , ..., 0. , 0. , 0. ], ..., [ 0. , 0. , 0. , ..., 0. , 0. , -0.02971415], [ 0. , 0. , 0. , ..., 0. , 0. , -0.1242566 ], [ 0. , 0. , 0. , ..., 0. , 0. , 0.11878401]]),) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62431/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62431/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62430	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62430/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62430/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62430/events	https://github.com/tensorflow/tensorflow/pull/62430	2,000,762,259	PR_kwDOArmXAs5f1dwm	62,430	TFLite GPU: fix certain tests for OSS users	{ "login": "looi", "id": 1855766, "node_id": "MDQ6VXNlcjE4NTU3NjY=", "avatar_url": "https://avatars.githubusercontent.com/u/1855766?v=4", "gravatar_id": "", "url": "https://api.github.com/users/looi", "html_url": "https://github.com/looi", "followers_url": "https://api.github.com/users/looi/followers", "following_url": "https://api.github.com/users/looi/following{/other_user}", "gists_url": "https://api.github.com/users/looi/gists{/gist_id}", "starred_url": "https://api.github.com/users/looi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/looi/subscriptions", "organizations_url": "https://api.github.com/users/looi/orgs", "repos_url": "https://api.github.com/users/looi/repos", "events_url": "https://api.github.com/users/looi/events{/privacy}", "received_events_url": "https://api.github.com/users/looi/received_events", "type": "User", "site_admin": false }	[ { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169365494, "node_id": "MDU6TGFiZWwxMTY5MzY1NDk0", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:M", "name": "size:M", "color": "adafea", "default": false, "description": "CL Change Size: Medium" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[]	2023-11-19T09:01:11	2023-11-20T07:17:17	2023-11-20T05:37:36	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62430", "html_url": "https://github.com/tensorflow/tensorflow/pull/62430", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62430.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62430.patch", "merged_at": "2023-11-20T05:37:36" }	After 3dc509f31848c7778dc68fabc59ab39c2e0d1e4a, cannot build for OSS users. Also, EXPECT_OK needs to be replaced with ASSERT_OK. @grantjensen	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62430/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62430/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62429	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62429/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62429/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62429/events	https://github.com/tensorflow/tensorflow/issues/62429	2,000,754,918	I_kwDOArmXAs53QRjm	62,429	Bug in tf.test.compute_gradient for tf.keras.layers.Activation	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @cheyennee ,\r\n\r\nI have replicated the reported behaviour with colab using TF v2.14, 2.15, and TF-nightly. Please find the [gist](https://colab.research.google.com/gist/Venkat6871/f18fd1e643b1a8eb6190a5e0a2e90488/62429_v2-14_2-15_nightly.ipynb) here for reference.\r\n\r\nThank you!" ]	2023-11-19T08:38:35	2024-01-30T10:37:36	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? The numerical gradient and theoretical gradient for `tf.keras.layers.Activation` is different, one is `0.5`, the other is `0`. ### Standalone code to reproduce the issue ```shell import tensorflow as tf activation = "relu" __input___0_tensor = tf.random.uniform([2, 7, 1, 1], minval=0, maxval=0, dtype=tf.float64) __input___0 = tf.identity(__input___0_tensor) Activation_class = tf.keras.layers.Activation(activation) layer = Activation_class inputs = __input___0 r = Activation_class(inputs) theoretical, numerical = tf.test.compute_gradient(Activation_class, [inputs]) print(theoretical) print(numerical) ``` ### Relevant log output ```shell (array([[0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.], [0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0., 0.]]),) (array([[0.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0.5, 0. , 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.5, 0. , 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.5, 0. , 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.5, 0. , 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.5, 0. , 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.5, 0. , 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.5, 0. ], [0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0. , 0.5]]),) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62429/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62429/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62428	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62428/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62428/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62428/events	https://github.com/tensorflow/tensorflow/pull/62428	2,000,720,756	PR_kwDOArmXAs5f1Vgi	62,428	[oneDNN]Implement portserver on Windows	{ "login": "mraunak", "id": 83710963, "node_id": "MDQ6VXNlcjgzNzEwOTYz", "avatar_url": "https://avatars.githubusercontent.com/u/83710963?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mraunak", "html_url": "https://github.com/mraunak", "followers_url": "https://api.github.com/users/mraunak/followers", "following_url": "https://api.github.com/users/mraunak/following{/other_user}", "gists_url": "https://api.github.com/users/mraunak/gists{/gist_id}", "starred_url": "https://api.github.com/users/mraunak/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/mraunak/subscriptions", "organizations_url": "https://api.github.com/users/mraunak/orgs", "repos_url": "https://api.github.com/users/mraunak/repos", "events_url": "https://api.github.com/users/mraunak/events{/privacy}", "received_events_url": "https://api.github.com/users/mraunak/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169365682, "node_id": "MDU6TGFiZWwxMTY5MzY1Njgy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:L", "name": "size:L", "color": "adafea", "default": false, "description": "CL Change Size: Large" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[]	2023-11-19T06:40:50	2023-11-22T20:39:37	2023-11-22T20:39:37	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62428", "html_url": "https://github.com/tensorflow/tensorflow/pull/62428", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62428.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62428.patch", "merged_at": "2023-11-22T20:39:37" }	This PR starts the portserver before running the Bazel tests related to TensorFlow on the Windows platform and stops the port server once the test runs are completed. The use of portserver during testing helps to avoid duplicity of ports attempted to be used by tests running in parallel. Porserver guides portpicker library to select ports The reference PR https://github.com/tensorflow/tensorflow/pull/62052 is the implementation of Portserver on Linux	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62428/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62428/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62427	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62427/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62427/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62427/events	https://github.com/tensorflow/tensorflow/issues/62427	2,000,685,608	I_kwDOArmXAs53QAoo	62,427	tf.keras.layers.ConvLSTM2D throws error when backprop	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@sachinprasadhs \r\nI was able to replicate the reported behaviour with TF2.13, TF2.14 and tf-nightly. Kindly find the [gist](https://colab.research.google.com/gist/tilakrayal/264de69376578a37cec37af078d8cee2/untitled1522.ipynb) for reference.\r\n" ]	2023-11-19T03:58:36	2023-12-26T19:44:56	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? API crashes in backprop. ### Standalone code to reproduce the issue ```shell filters = 2 kernel_size_0 = 4 kernel_size_1 = 2 kernel_size = [kernel_size_0,kernel_size_1,] strides_0 = 1 strides_1 = 1 strides = [strides_0,strides_1,] padding = "valid" data_format = "channels_first" dilation_rate_0 = 1 dilation_rate_1 = 1 dilation_rate = [dilation_rate_0,dilation_rate_1,] activation = "tanh" recurrent_activation = "hard_sigmoid" use_bias = False kernel_initializer = None recurrent_initializer = None bias_initializer = "zeros" unit_forget_bias = True kernel_regularizer = None recurrent_regularizer = None bias_regularizer = None activity_regularizer = None kernel_constraint = None recurrent_constraint = None bias_constraint = None return_sequences = False return_state = False go_backwards = True stateful = False dropout = 0.0 recurrent_dropout = 0.0 __input___0_tensor = tf.random.uniform([2, 2, 2, 5, 1], minval=1.7518786460769893, maxval=2.8945805380363145, dtype=tf.float32) __input___0 = tf.identity(__input___0_tensor) ConvLSTM2D_class = tf.keras.layers.ConvLSTM2D(filters, kernel_size, strides=strides, padding=padding, data_format=data_format, dilation_rate=dilation_rate, activation=activation, recurrent_activation=recurrent_activation, use_bias=use_bias, kernel_initializer=kernel_initializer, recurrent_initializer=recurrent_initializer, bias_initializer=bias_initializer, unit_forget_bias=unit_forget_bias, kernel_regularizer=kernel_regularizer, recurrent_regularizer=recurrent_regularizer, bias_regularizer=bias_regularizer, activity_regularizer=activity_regularizer, kernel_constraint=kernel_constraint, recurrent_constraint=recurrent_constraint, bias_constraint=bias_constraint, return_sequences=return_sequences, return_state=return_state, go_backwards=go_backwards, stateful=stateful, dropout=dropout, recurrent_dropout=recurrent_dropout) layer = ConvLSTM2D_class inputs = __input___0 with tf.GradientTape() as g: g.watch(inputs) res = layer(inputs) print(res.shape) grad = g.jacobian(res, inputs) # Error ``` ### Relevant log output ```shell (2, 2, 2, 0) --------------------------------------------------------------------------- ValueError Traceback (most recent call last) <ipython-input-15-75a544b673ad> in <cell line: 57>() 55 res = layer(inputs) 56 print(res.shape) ---> 57 grad = g.jacobian(res, inputs) # Error 58 # print(grad) 1 frames /usr/local/lib/python3.10/dist-packages/tensorflow/python/framework/ops.py in set_shape(self, shape) 534 def set_shape(self, shape) -> None: 535 if not self.shape.is_compatible_with(shape): --> 536 raise ValueError(f"Tensor's shape {self.shape} is not compatible " 537 f"with supplied shape {shape}.") 538 ValueError: Tensor's shape (2, 2, 2, 0, 1, 2, 2, 5, 1) is not compatible with supplied shape (2, 2, 2, 0, 2, 2, 2, 5, 1). ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62427/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62427/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62426	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62426/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62426/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62426/events	https://github.com/tensorflow/tensorflow/issues/62426	2,000,674,127	I_kwDOArmXAs53P91P	62,426	tf.keras.layers.Conv3DTranspose throws error when backprop	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Same problem can be found in tf.keras.layers.LocallyConnected2D. Here is the repo code:\r\n```\r\nfilters = 2048\r\nkernel_size_0 = 3\r\nkernel_size_1 = 3\r\nkernel_size = [kernel_size_0,kernel_size_1,]\r\nstrides_0 = 1\r\nstrides_1 = 1\r\nstrides = [strides_0,strides_1,]\r\npadding = \"valid\"\r\ndata_format = None\r\nactivation = \"linear\"\r\nuse_bias = True\r\nkernel_initializer = None\r\nbias_initializer = None\r\nkernel_regularizer = None\r\nbias_regularizer = None\r\nactivity_regularizer = None\r\nkernel_constraint = None\r\nbias_constraint = None\r\nimplementation = 1\r\n__input___0_tensor = tf.random.uniform([1, 6, 10, 1], minval=1.1, maxval=10, dtype=tf.float64)\r\n__input___0 = tf.identity(__input___0_tensor)\r\nLocallyConnected2D_class = tf.keras.layers.LocallyConnected2D(filters, kernel_size, strides=strides, padding=padding, data_format=data_format, activation=activation, use_bias=use_bias, kernel_initializer=kernel_initializer, bias_initializer=bias_initializer, kernel_regularizer=kernel_regularizer, bias_regularizer=bias_regularizer, activity_regularizer=activity_regularizer, kernel_constraint=kernel_constraint, bias_constraint=bias_constraint, implementation=implementation)\r\n\r\nlayer = LocallyConnected2D_class\r\ninputs = __input___0\r\n\r\nwith tf.GradientTape() as g:\r\n g.watch(inputs)\r\n res = layer(inputs)\r\nprint(res.shape)\r\ngrad = g.jacobian(res, inputs)\r\nprint(grad)\r\n```", "Hi @cheyennee ,\r\n\r\nI have tested the given code on colab and its working fine.Please refer attached [gist](https://colab.sandbox.google.com/gist/SuryanarayanaY/ef3f074245957fa1ce975893ed897f39/62426.ipynb). \r\n\r\nPlease note that I have reduced the no of filters due to Memory constraints but it should not affect the reported behaviour.\r\n\r\nCould you please verify the behaviour attached. Can you confirm whether the issue with Windows Package as it will download intel package?\r\n\r\nThanks!", "Hi @SuryanarayanaY,\r\nIt appears that the number of filters does make a difference 😂. In your [gist](https://colab.research.google.com/gist/SuryanarayanaY/ef3f074245957fa1ce975893ed897f39/62426.ipynb#scrollTo=1chUD-vIX7Dg), where the number of filters is set to 40, there are no crashes. However, I repo above code in colab, in the provided [gist](https://colab.research.google.com/drive/1zOQm4FfPCepRdBH9M3-ef6KV8_wbF3jA), where the number of filters is increased to 1792, it crashes, and the error message suggests a potential OOM issue. ", "Same problem can be found in tf.keras.layers.UpSampling2D. Here is the repo code:\r\n```\r\nimport tensorflow as tf\r\n\r\nsize_0 = 224\r\nsize_1 = 224\r\nsize = [size_0,size_1, ]\r\ndata_format = None\r\ninterpolation = \"nearest\"\r\n__input___0_tensor = tf.random.uniform([1, 8, 8, 1], minval=0.0, maxval=2.84955773162074, dtype=tf.float64)\r\n__input___0 = tf.identity(__input___0_tensor)\r\nUpSampling2D_class = tf.keras.layers.UpSampling2D(size=size, data_format=data_format, interpolation=interpolation)\r\n\r\n\r\nlayer = UpSampling2D_class\r\ninputs = __input___0\r\n\r\nwith tf.GradientTape() as g:\r\n g.watch(inputs)\r\n res = layer(inputs)\r\nprint(res.shape)\r\ngrad = g.jacobian(res, inputs)\r\nprint(grad)\r\n```\r\noutput is:\r\n```\r\n(1, 1792, 1792, 1)\r\n---------------------------------------------------------------------------\r\nResourceExhaustedError Traceback (most recent call last)\r\n[<ipython-input-1-8abc242ca0fd>](https://localhost:8080/#) in <cell line: 20>()\r\n 18 res = layer(inputs)\r\n 19 print(res.shape)\r\n---> 20 grad = g.jacobian(res, inputs)\r\n 21 print(grad)\r\n\r\n3 frames\r\n[/usr/local/lib/python3.10/dist-packages/tensorflow/python/eager/execute.py](https://localhost:8080/#) in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name)\r\n 58 for t in inputs\r\n 59 ]\r\n---> 60 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name,\r\n 61 inputs, attrs, num_outputs)\r\n 62 except core._NotOkStatusException as e:\r\n\r\nResourceExhaustedError: Graph execution error:\r\n\r\nDetected at node gradient_tape/UnsortedSegmentSum/pfor/UnsortedSegmentSum defined at (most recent call last):\r\n File \"/usr/lib/python3.10/runpy.py\", line 196, in _run_module_as_main\r\n\r\n File \"/usr/lib/python3.10/runpy.py\", line 86, in _run_code\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py\", line 37, in <module>\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/traitlets/config/application.py\", line 992, in launch_instance\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/ipykernel/kernelapp.py\", line 619, in start\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/tornado/platform/asyncio.py\", line 195, in start\r\n\r\n File \"/usr/lib/python3.10/asyncio/base_events.py\", line 603, in run_forever\r\n\r\n File \"/usr/lib/python3.10/asyncio/base_events.py\", line 1909, in _run_once\r\n\r\n File \"/usr/lib/python3.10/asyncio/events.py\", line 80, in _run\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py\", line 685, in <lambda>\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py\", line 738, in _run_callback\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/tornado/gen.py\", line 825, in inner\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/tornado/gen.py\", line 786, in run\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py\", line 361, in process_one\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/tornado/gen.py\", line 234, in wrapper\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py\", line 261, in dispatch_shell\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/tornado/gen.py\", line 234, in wrapper\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py\", line 539, in execute_request\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/tornado/gen.py\", line 234, in wrapper\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py\", line 302, in do_execute\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/ipykernel/zmqshell.py\", line 539, in run_cell\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 2975, in run_cell\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3030, in _run_cell\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/IPython/core/async_helpers.py\", line 78, in _pseudo_sync_runner\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3257, in run_cell_async\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3473, in run_ast_nodes\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py\", line 3553, in run_code\r\n\r\n File \"<ipython-input-1-8abc242ca0fd>\", line 20, in <cell line: 20>\r\n\r\n File \"/usr/local/lib/python3.10/dist-packages/tensorflow/python/ops/parallel_for/control_flow_ops.py\", line 212, in f\r\n\r\nOOM when allocating tensor with shape[10312216477696] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc\r\n\t [[{{node gradient_tape/UnsortedSegmentSum/pfor/UnsortedSegmentSum}}]]\r\nHint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info. This isn't available when running in Eager mode.\r\n [Op:__inference_f_162]\r\n```\r\nIt may be that the size is too large and the memory is overflowing.", "Same problem can be found in tf.keras.layers.ZeroPadding3D. Here is the repo code:\r\n```\r\npadding = 16\r\ndata_format = None\r\n__input___0_tensor = tf.random.uniform([1, 1, 2, 2, 3], minval=0.8510533546655319,maxval=3.0,dtype=tf.float64)\r\n__input___0 = tf.identity(__input___0_tensor)\r\nZeroPadding3D_class = tf.keras.layers.ZeroPadding3D(padding=padding,data_format=data_format)\r\n\r\nlayer = ZeroPadding3D_class\r\ninputs = __input___0\r\n\r\nwith tf.GradientTape() as g:\r\n g.watch(inputs)\r\n res = layer(inputs)\r\nprint(res.shape)\r\ngrad = g.jacobian(res, inputs)\r\nprint(grad)\r\n```", "> Hi @SuryanarayanaY, It appears that the number of filters does make a difference 😂. In your [gist](https://colab.research.google.com/gist/SuryanarayanaY/ef3f074245957fa1ce975893ed897f39/62426.ipynb#scrollTo=1chUD-vIX7Dg), where the number of filters is set to 40, there are no crashes. However, I repo above code in colab, in the provided [gist](https://colab.research.google.com/drive/1zOQm4FfPCepRdBH9M3-ef6KV8_wbF3jA), where the number of filters is increased to 1792, it crashes, and the error message suggests a potential OOM issue.\r\n\r\nHi @cheyennee , OOM errors are not problem with Tensorflow and with low number of filters it is indeed working.\r\n\r\n\r\nOh Sorry,I missed this from your logs.\r\n\r\n```\r\nOOM when allocating tensor with shape[50982027264] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc\r\n\t [[{{node gradient_tape/UnsortedSegmentSum/pfor/UnsortedSegmentSum}}]]\r\nHint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info. This isn't available when running in Eager mode.\r\n [Op:__inference_f_52731]\r\n```\r\nThis is OOM error and not a problem with Tensorflow. With higher input size this is intended.", "Hi @SuryanarayanaY , I understand. It appears that this issue is related to the device I am currently using. I'll reduce the no of filters. Thank you!", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62426\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62426\">No</a>\n" ]	2023-11-19T03:02:23	2023-11-21T02:32:04	2023-11-21T02:32:00	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? During backpropagation, the API crashes. Based on the error message, it appears to be indicative of an OOM situation. ### Standalone code to reproduce the issue ```shell filters = 1792 kernel_size_0 = 3 kernel_size_1 = 3 kernel_size_2 = 3 kernel_size = [kernel_size_0,kernel_size_1,kernel_size_2,] strides_0 = 1 strides_1 = 1 strides_2 = 1 strides = [strides_0,strides_1,strides_2,] padding = "valid" output_padding = None data_format = "channels_last" dilation_rate_0 = 1 dilation_rate_1 = 1 dilation_rate_2 = 1 dilation_rate = [dilation_rate_0,dilation_rate_1,dilation_rate_2,] activation = "linear" use_bias = True kernel_initializer = None bias_initializer = None kernel_regularizer = None bias_regularizer = None activity_regularizer = None kernel_constraint = None bias_constraint = None __input___0_tensor = tf.random.uniform([2, 5, 1, 1, 1], minval=0, maxval=1, dtype=tf.float32) __input___0 = tf.identity(__input___0_tensor) Conv3DTranspose_class = tf.keras.layers.Conv3DTranspose(filters, kernel_size, strides=strides, padding=padding, output_padding=output_padding, data_format=data_format, dilation_rate=dilation_rate, activation=activation, use_bias=use_bias, kernel_initializer=kernel_initializer, bias_initializer=bias_initializer, kernel_regularizer=kernel_regularizer, bias_regularizer=bias_regularizer, activity_regularizer=activity_regularizer, kernel_constraint=kernel_constraint, bias_constraint=bias_constraint) layer = Conv3DTranspose_class inputs = __input___0 with tf.GradientTape() as g: g.watch(inputs) res = layer(inputs) print(res.shape) grad = g.jacobian(res, inputs) # Error ``` ### Relevant log output ```shell (2, 7, 3, 3, 1792) --------------------------------------------------------------------------- ResourceExhaustedError Traceback (most recent call last) <ipython-input-6-15c96e537b1f> in <cell line: 47>() 45 res = layer(inputs) 46 print(res.shape) # (1, 5, 6, 2) ---> 47 grad = g.jacobian(res, inputs) # Error 3 frames /usr/local/lib/python3.10/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name) 58 for t in inputs 59 ] ---> 60 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name, 61 inputs, attrs, num_outputs) 62 except core._NotOkStatusException as e: ResourceExhaustedError: Graph execution error: Detected at node gradient_tape/UnsortedSegmentSum/pfor/UnsortedSegmentSum defined at (most recent call last): File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main File "/usr/lib/python3.10/runpy.py", line 86, in _run_code File "/usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py", line 37, in <module> File "/usr/local/lib/python3.10/dist-packages/traitlets/config/application.py", line 992, in launch_instance File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelapp.py", line 619, in start File "/usr/local/lib/python3.10/dist-packages/tornado/platform/asyncio.py", line 195, in start File "/usr/lib/python3.10/asyncio/base_events.py", line 603, in run_forever File "/usr/lib/python3.10/asyncio/base_events.py", line 1909, in _run_once File "/usr/lib/python3.10/asyncio/events.py", line 80, in _run File "/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py", line 685, in <lambda> File "/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py", line 738, in _run_callback File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 825, in inner File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 786, in run File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 361, in process_one File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 261, in dispatch_shell File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 539, in execute_request File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py", line 302, in do_execute File "/usr/local/lib/python3.10/dist-packages/ipykernel/zmqshell.py", line 539, in run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 2975, in run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3030, in _run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/async_helpers.py", line 78, in _pseudo_sync_runner File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3257, in run_cell_async File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3473, in run_ast_nodes File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3553, in run_code File "<ipython-input-6-15c96e537b1f>", line 47, in <cell line: 47> File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/ops/parallel_for/control_flow_ops.py", line 212, in f OOM when allocating tensor with shape[50982027264] and type float on /job:localhost/replica:0/task:0/device:GPU:0 by allocator GPU_0_bfc [[{{node gradient_tape/UnsortedSegmentSum/pfor/UnsortedSegmentSum}}]] Hint: If you want to see a list of allocated tensors when OOM happens, add report_tensor_allocations_upon_oom to RunOptions for current allocation info. This isn't available when running in Eager mode. [Op:__inference_f_52731] ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62426/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62426/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62425	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62425/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62425/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62425/events	https://github.com/tensorflow/tensorflow/issues/62425	2,000,445,676	I_kwDOArmXAs53PGDs	62,425	tf.keras.layers.Conv3D throws error when backprop	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@cheyennee I was able to replicate the issue on colab, please find the gist [here](https://colab.research.google.com/gist/sushreebarsa/b2b68382c6bd9682aff75bb1c40dc880/62425.ipynb#scrollTo=M60O0nHx2Qdj). Could you either increase the strides or decrease the kernel_size to avoid such an issue?\r\nThank you!", "Hi, @sushreebarsa. Decreasing the kernel_size can avoid such an issue. But shouldn't this error message be set to be more user-friendly, so that the user knows how to adjust the parameters appropriately?" ]	2023-11-18T14:34:08	2023-12-26T23:10:19	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? When I try to do back prop on tf.keras.layers.Conv3D layer, it throws error. I think this bug is related to tensorflow because the error message contains tensorflow backprop ops. ### Standalone code to reproduce the issue ```shell filters = 1 kernel_size_0 = 4 kernel_size_1 = 2 kernel_size_2 = 1 kernel_size = [kernel_size_0, kernel_size_1,kernel_size_2, ] strides_0 = 1 strides_1 = 1 strides_2 = 1 strides = [strides_0, strides_1, strides_2, ] padding = "valid" data_format = "channels_last" dilation_rate_0 = 1 dilation_rate_1 = 1 dilation_rate_2 = 1 dilation_rate = [dilation_rate_0,dilation_rate_1,dilation_rate_2,] groups = 1 activation = "relu" use_bias = True kernel_initializer = None bias_initializer = None kernel_regularizer = None bias_regularizer = None activity_regularizer = None kernel_constraint = None bias_constraint = None x = tf.random.uniform([1, 3, 3, 3, 1], minval=0.7301831337248943, maxval=3.0, dtype=tf.float32) layer = tf.keras.layers.Conv3D(filters, kernel_size, strides=strides, padding=padding, data_format=data_format, dilation_rate=dilation_rate, groups=groups, activation=activation, use_bias=use_bias, kernel_initializer=kernel_initializer, bias_initializer=bias_initializer, kernel_regularizer=kernel_regularizer, bias_regularizer=bias_regularizer, activity_regularizer=activity_regularizer, kernel_constraint=kernel_constraint, bias_constraint=bias_constraint) with tf.GradientTape() as g: g.watch(x) res = layer(x) grad = g.jacobian(res, x) # Error ``` ### Relevant log output ```shell InvalidArgumentError Traceback (most recent call last) <ipython-input-16-7a29e2e0c3b9> in <cell line: 34>() 32 res = layer(x) 33 print(res.shape) ---> 34 grad = g.jacobian(res, x) # Error 3 frames /usr/local/lib/python3.10/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name) 58 for t in inputs 59 ] ---> 60 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name, 61 inputs, attrs, num_outputs) 62 except core._NotOkStatusException as e: InvalidArgumentError: Graph execution error: Detected at node gradient_tape/Conv3DBackpropInputV2/pfor/TensorArrayV2Stack/TensorListStack defined at (most recent call last): File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main File "/usr/lib/python3.10/runpy.py", line 86, in _run_code File "/usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py", line 37, in <module> File "/usr/local/lib/python3.10/dist-packages/traitlets/config/application.py", line 992, in launch_instance File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelapp.py", line 619, in start File "/usr/local/lib/python3.10/dist-packages/tornado/platform/asyncio.py", line 195, in start File "/usr/lib/python3.10/asyncio/base_events.py", line 603, in run_forever File "/usr/lib/python3.10/asyncio/base_events.py", line 1909, in _run_once File "/usr/lib/python3.10/asyncio/events.py", line 80, in _run File "/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py", line 685, in <lambda> File "/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py", line 738, in _run_callback File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 825, in inner File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 786, in run File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 361, in process_one File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 261, in dispatch_shell File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 539, in execute_request File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py", line 302, in do_execute File "/usr/local/lib/python3.10/dist-packages/ipykernel/zmqshell.py", line 539, in run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 2975, in run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3030, in _run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/async_helpers.py", line 78, in _pseudo_sync_runner File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3257, in run_cell_async File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3473, in run_ast_nodes File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3553, in run_code File "<ipython-input-16-7a29e2e0c3b9>", line 34, in <cell line: 34> File "/usr/local/lib/python3.10/dist-packages/tensorflow/python/ops/parallel_for/control_flow_ops.py", line 212, in f Tried to stack elements of an empty list with non-fully-defined element_shape: [?,?,?,?,?] [[{{node gradient_tape/Conv3DBackpropInputV2/pfor/TensorArrayV2Stack/TensorListStack}}]] [Op:__inference_f_4984] ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62425/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62425/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62424	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62424/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62424/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62424/events	https://github.com/tensorflow/tensorflow/issues/62424	2,000,399,910	I_kwDOArmXAs53O64m	62,424	Bug in backward gradient for tf.keras.layers.Conv2DTranspose	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" } ]	closed	false	{ "login": "Venkat6871", "id": 147127861, "node_id": "U_kgDOCMT-NQ", "avatar_url": "https://avatars.githubusercontent.com/u/147127861?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Venkat6871", "html_url": "https://github.com/Venkat6871", "followers_url": "https://api.github.com/users/Venkat6871/followers", "following_url": "https://api.github.com/users/Venkat6871/following{/other_user}", "gists_url": "https://api.github.com/users/Venkat6871/gists{/gist_id}", "starred_url": "https://api.github.com/users/Venkat6871/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Venkat6871/subscriptions", "organizations_url": "https://api.github.com/users/Venkat6871/orgs", "repos_url": "https://api.github.com/users/Venkat6871/repos", "events_url": "https://api.github.com/users/Venkat6871/events{/privacy}", "received_events_url": "https://api.github.com/users/Venkat6871/received_events", "type": "User", "site_admin": false }	[ { "login": "Venkat6871", "id": 147127861, "node_id": "U_kgDOCMT-NQ", "avatar_url": "https://avatars.githubusercontent.com/u/147127861?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Venkat6871", "html_url": "https://github.com/Venkat6871", "followers_url": "https://api.github.com/users/Venkat6871/followers", "following_url": "https://api.github.com/users/Venkat6871/following{/other_user}", "gists_url": "https://api.github.com/users/Venkat6871/gists{/gist_id}", "starred_url": "https://api.github.com/users/Venkat6871/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Venkat6871/subscriptions", "organizations_url": "https://api.github.com/users/Venkat6871/orgs", "repos_url": "https://api.github.com/users/Venkat6871/repos", "events_url": "https://api.github.com/users/Venkat6871/events{/privacy}", "received_events_url": "https://api.github.com/users/Venkat6871/received_events", "type": "User", "site_admin": false } ]	null	[ "close as it fixed in tf v2.14, while can be repro in tf v2.9 ", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62424\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62424\">No</a>\n" ]	2023-11-18T12:26:28	2023-11-18T13:02:12	2023-11-18T13:02:08	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? The shape of res_backward is (1, 5, 6, 2), the shape of res_forward is (1, 5, 6, 2), the shape of grad_jvp is (1, 5, 6, 2), while the shape of backward gradient is (1, 5, 6, 2, 1, 5, 6, 2). I think both the values and shapes of grad_backward and grad_jvp should be the same. ### Standalone code to reproduce the issue ```shell filters = 2 kernel_size_0 = 3 kernel_size_1 = 3 kernel_size = [kernel_size_0,kernel_size_1, ] strides_0 = 1 strides_1 = 1 strides = [ strides_0, strides_1, ] padding = "same" output_padding = None data_format = "channels_last" dilation_rate_0 = 2 dilation_rate_1 = 2 dilation_rate = [dilation_rate_0,dilation_rate_1,] activation = "linear" use_bias = True kernel_initializer = None bias_initializer = None kernel_regularizer = None bias_regularizer = None activity_regularizer = None kernel_constraint = None bias_constraint = None __input___0_tensor = tf.random.uniform([1, 5, 6, 1], minval=-2, maxval=2, dtype=tf.float32) __input___0 = tf.identity(__input___0_tensor) Conv2DTranspose_class = tf.keras.layers.Conv2DTranspose(filters, kernel_size, strides=strides, padding=padding, output_padding=output_padding, data_format=data_format, dilation_rate=dilation_rate, activation=activation, use_bias=use_bias, kernel_initializer=kernel_initializer, bias_initializer=bias_initializer, kernel_regularizer=kernel_regularizer, bias_regularizer=bias_regularizer, activity_regularizer=activity_regularizer, kernel_constraint=kernel_constraint, bias_constraint=bias_constraint, dtype=tf.float32) layer = Conv2DTranspose_class inputs = __input___0 with tf.GradientTape(persistent=True, ) as g: g.watch(inputs) res_backward = layer(inputs) grad_backward = g.jacobian(res_backward, res_backward) print("res_backward:", res_backward) print("grad_backward:", grad_backward) tangents = tf.constant(1., dtype=tf.float32, shape=[1, 5, 6, 1]) with tf.autodiff.ForwardAccumulator(inputs, tangents) as acc: res_forward = layer(inputs) grad_jvp = acc.jvp(res_forward) print("res_forward:", res_forward) print("grad_forward", grad_jvp) ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62424/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62424/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62423	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62423/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62423/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62423/events	https://github.com/tensorflow/tensorflow/issues/62423	2,000,362,561	I_kwDOArmXAs53OxxB	62,423	Forward AD threw error in tf.autodiff.ForwardAccumulator for tf.keras.layers.AveragePooling3D, but backward AD succeeded with same input	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@cheyennee,\r\nLooks like this is duplicate of issue [#56833](https://github.com/tensorflow/tensorflow/issues/56833). Could you please close this issue, since it is already being tracked there? Thank you!", "Hi, @tilakrayal ,\r\nIt is not a duplicate of issue #56833. The provided code snippet triggers an error only when padding=\"valid\" is used, not when padding=\"same\". Additionally, in line with the details outlined in #56833, I have already configured the data_format parameter.", "@cheyennee,\r\nI was able to replicate the reported behaviour with TF2.14 and tf-nightly as well. Attached [gist](https://colab.research.google.com/gist/tilakrayal/c6b3af27249daeee7c178c4bb05628e3/untitled1521.ipynb) for reference.\r\nCurrently we are investigating the issue & will deep dive into the issue and provide the root-cause for the same. Thank you!" ]	2023-11-18T10:28:55	2023-12-26T23:09:19	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Forward AD threw error in tf.autodiff.ForwardAccumulator for tf.keras.layers.AveragePooling3D, but backward AD succeeded with same input. ### Standalone code to reproduce the issue ```shell import tensorflow as tf pool_size = [2,1,1] strides = [2,2,2] padding = "valid" data_format = "channels_last" input = tf.constant(0.895205,shape=[1,1,1,1,1], dtype=tf.float32) layer = tf.keras.layers.AveragePooling3D(pool_size=pool_size, strides=strides, padding=padding, data_format=data_format, ) with tf.GradientTape(persistent=True, ) as g: g.watch(input) res_backward = layer(input) grad_backward = g.jacobian(res_backward,res_backward) print("res_backward:",res_backward) print("grad_backward:",grad_backward) tangents = tf.constant(1.,dtype=tf.float32,shape=[1,1,1,1,1]) with tf.autodiff.ForwardAccumulator(input, tangents) as acc: res_forward = layer(input) grad_jvp = acc.jvp(res_forward) print("res_forward:", res_forward) print("grad_forward", grad_jvp) ``` ### Relevant log output ```shell res_backward: tf.Tensor([], shape=(1, 0, 1, 1, 1), dtype=float32) grad_backward: tf.Tensor([], shape=(1, 0, 1, 1, 1, 1, 0, 1, 1, 1), dtype=float32) --------------------------------------------------------------------------- ValueError Traceback (most recent call last) <ipython-input-3-c1962478ba56> in <cell line: 18>() 17 tangents = tf.constant(1.,dtype=tf.float32,shape=[1,1,1,1,1]) 18 with tf.autodiff.ForwardAccumulator(input, tangents) as acc: ---> 19 res_forward = layer(input) 20 grad_jvp = acc.jvp(res_forward) 21 print("res_forward:", res_forward) 1 frames /usr/local/lib/python3.10/dist-packages/tensorflow/python/framework/ops.py in _create_c_op(graph, node_def, inputs, control_inputs, op_def, extract_traceback) 1019 except errors.InvalidArgumentError as e: 1020 # Convert to ValueError for backwards compatibility. -> 1021 raise ValueError(e.message) 1022 1023 # Record the current Python stack trace as the creating stacktrace of this ValueError: Exception encountered when calling layer 'average_pooling3d_2' (type AveragePooling3D). Negative dimension size caused by subtracting 2 from 1 for '{{node gradient_tape/gradient_tape/AvgPool3D}} = AvgPool3D[T=DT_FLOAT, data_format="NDHWC", ksize=[1, 2, 1, 1, 1], padding="VALID", strides=[1, 2, 2, 2, 1]](tangents)' with input shapes: [1,1,1,1,1]. Call arguments received by layer 'average_pooling3d_2' (type AveragePooling3D): • inputs=tf.Tensor(shape=(1, 1, 1, 1, 1), dtype=float32) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62423/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62423/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62422	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62422/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62422/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62422/events	https://github.com/tensorflow/tensorflow/issues/62422	2,000,325,567	I_kwDOArmXAs53Oou_	62,422	Forward AD threw error in tf.autodiff.ForwardAccumulator for tf.keras.layers.AveragePooling2D, but backward AD succeeded with same input	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @cheyennee ,\r\n\r\nI have replicated the reported behaviour with TF2.14 and tf-nightly as well. Attached [gist](https://colab.sandbox.google.com/gist/SuryanarayanaY/6d5f658f11168b4c2a2581078dec3280/62422.ipynb) for reference.\r\n\r\nWill dig more into for the root cause and come back. Thanks!" ]	2023-11-18T08:44:06	2024-03-08T07:24:25	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Forward AD threw error in tf.autodiff.ForwardAccumulator for tf.keras.layers.AveragePooling2D, but backward AD succeeded with same input ### Standalone code to reproduce the issue ```shell pool_size_0 = 16 pool_size_1 = 3 pool_size = [pool_size_0,pool_size_1, ] strides = None padding = "valid" data_format = None __input___0_tensor = tf.random.uniform([2, 32, 1, 1], minval=0, maxval=2, dtype=tf.float32) __input___0 = tf.identity(__input___0_tensor) inputs = __input___0 layer = tf.keras.layers.AveragePooling2D(pool_size=pool_size, strides=strides, padding=padding, data_format=data_format, dtype=tf.float64) with tf.GradientTape(persistent=True, ) as g: g.watch(inputs) res_backward = layer(inputs) grad_backward = g.jacobian(res_backward, res_backward) print("res_backward:", res_backward) print("grad_backward:", grad_backward) tangents = tf.constant(1., dtype=tf.float32, shape=[2, 32, 1, 1]) with tf.autodiff.ForwardAccumulator(inputs, tangents) as acc: res_forward = layer(inputs) grad_jvp = acc.jvp(res_forward) print("res_forward:", res_forward) print("grad_forward", grad_jvp) ``` ### Relevant log output ```shell res_backward: tf.Tensor([], shape=(2, 2, 0, 1), dtype=float64) grad_backward: tf.Tensor([], shape=(2, 2, 0, 1, 2, 2, 0, 1), dtype=float64) --------------------------------------------------------------------------- ValueError Traceback (most recent call last) <ipython-input-43-1e65b8d83cd7> in <cell line: 22>() 21 22 with tf.autodiff.ForwardAccumulator(inputs, tangents) as acc: ---> 23 res_forward = layer(inputs) 24 grad_jvp = acc.jvp(res_forward) 25 print("res_forward:", res_forward) 1 frames /usr/local/lib/python3.10/dist-packages/tensorflow/python/framework/ops.py in _create_c_op(graph, node_def, inputs, control_inputs, op_def, extract_traceback) 1019 except errors.InvalidArgumentError as e: 1020 # Convert to ValueError for backwards compatibility. -> 1021 raise ValueError(e.message) 1022 1023 # Record the current Python stack trace as the creating stacktrace of this ValueError: Exception encountered when calling layer 'average_pooling2d_12' (type AveragePooling2D). Negative dimension size caused by subtracting 3 from 1 for '{{node gradient_tape/gradient_tape/AvgPool}} = AvgPool[T=DT_DOUBLE, data_format="NHWC", ksize=[1, 16, 3, 1], padding="VALID", strides=[1, 16, 3, 1]](tangents)' with input shapes: [2,32,1,1]. Call arguments received by layer 'average_pooling2d_12' (type AveragePooling2D): • inputs=tf.Tensor(shape=(2, 32, 1, 1), dtype=float64) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62422/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62422/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62421	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62421/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62421/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62421/events	https://github.com/tensorflow/tensorflow/issues/62421	2,000,297,899	I_kwDOArmXAs53Oh-r	62,421	tf.keras.layers.AlphaDropout crash when noise_shape is a tensor	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@cheyennee I was able to replicate the issue reported [here](https://colab.research.google.com/gist/sushreebarsa/3006e9f5c94eb8adbad8f36ccb9010db/62421.ipynb). Could you please use a fixed value for noise_shape instead of a tensor as a workaround?\r\nThank you!", "Hi, @sushreebarsa. Using a fixed value for `noise_shape` works well. But I think it would be better if the documentation include details about the usage of `noise_shape` or that the error message generated `by tf.keras.layers.AlphaDropout` be improved to offer clearer guidance in the event of a crash." ]	2023-11-18T07:53:09	2023-12-26T23:11:20	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? In the documentation, there is no information provided for the `noise_shape` parameter. When the value of `noise_shape` is set as a tensor, the API crashes. It is recommended that the documentation include details about the usage of `noise_shape` or that the error message generated by `tf.keras.layers.AlphaDropout` be improved to offer clearer guidance in the event of a crash. ### Standalone code to reproduce the issue ```shell import tensorflow as tf __input___0_tensor = tf.random.uniform([3, 2, 3], minval=-1.0, maxval=0.04343286078799158, dtype=tf.float32) __input___0 = tf.identity(__input___0_tensor) noise_shape_tensor = tf.random.uniform([3, 2, 3, 1], minval=-2, maxval=2, dtype=tf.float32) noise_shape = tf.identity(noise_shape_tensor) rate = 0.5 seed = None alpha_dropout_layer = tf.keras.layers.AlphaDropout(rate, noise_shape=noise_shape, seed=seed) output = alpha_dropout_layer(__input___0) print(output) ``` ### Relevant log output ```shell ValueError Traceback (most recent call last) <ipython-input-30-fe6bb6d4ede4> in <cell line: 9>() 7 seed = None 8 alpha_dropout_layer = tf.keras.layers.AlphaDropout(rate, noise_shape=noise_shape, seed=seed) ----> 9 output = alpha_dropout_layer(__input___0) 10 print(output) 1 frames /usr/local/lib/python3.10/dist-packages/tensorflow/python/framework/ops.py in __bool__(self) 300 301 def __bool__(self) -> bool: --> 302 return bool(self._numpy()) 303 304 __nonzero__ = __bool__ ValueError: Exception encountered when calling layer 'alpha_dropout_12' (type AlphaDropout). The truth value of an array with more than one element is ambiguous. Use a.any() or a.all() Call arguments received by layer 'alpha_dropout_12' (type AlphaDropout): • inputs=tf.Tensor(shape=(3, 2, 3), dtype=float32) • training=None ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62421/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62421/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62420	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62420/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62420/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62420/events	https://github.com/tensorflow/tensorflow/issues/62420	2,000,278,192	I_kwDOArmXAs53OdKw	62,420	Wrong gradient calculated for API tf.keras.layers.Add	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @cheyennee ,\r\n\r\nI was able to reproduce the issue on colab using TF v2.14, 2.15, and TF-nightly. Please find the [gist](https://colab.research.google.com/gist/Venkat6871/9def4c1e50ace380e2e05d32980b89ab/62420_2-14_2-15-nightly_.ipynb) here for reference.\r\n\r\nThank you!" ]	2023-11-18T06:48:12	2023-11-28T16:06:39	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Forward mode differentiation for the case is inconsistent with the gradient calculated in reverse mode. They shoule be the same. ### Standalone code to reproduce the issue ```shell import tensorflow as tf inputs_0_tensor = tf.random.uniform([], minval=0.3, maxval=1, dtype=tf.float32) inputs_0 = tf.identity(inputs_0_tensor) inputs_1_tensor = tf.random.uniform([], minval=1.1, maxval=10,dtype=tf.float32) inputs_1 = tf.identity(inputs_1_tensor) inputs = [inputs_0, inputs_1] layer = tf.keras.layers.Add() with tf.GradientTape(persistent=True, ) as g: g.watch(inputs) res_backward = layer(inputs) grad_backward = g.jacobian(res_backward, res_backward) print("res_backward:", res_backward) print("grad_backward:", grad_backward) tangent_1 = tf.constant(1,dtype=tf.float32) tangent_2 = tf.constant(1,dtype=tf.float32) tangents = [tangent_1, tangent_2] with tf.autodiff.ForwardAccumulator(inputs, tangents) as acc: res_forward = layer(inputs) grad_jvp = acc.jvp(res_forward) print("res_forward:", res_forward) print("grad_forward", grad_jvp) ``` ### Relevant log output ```shell res_backward: tf.Tensor(9.794308, shape=(), dtype=float32) grad_backward: tf.Tensor(1.0, shape=(), dtype=float32) res_forward: tf.Tensor(9.794308, shape=(), dtype=float32) grad_forward tf.Tensor(2.0, shape=(), dtype=float32) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62420/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62420/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62419	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62419/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62419/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62419/events	https://github.com/tensorflow/tensorflow/issues/62419	1,999,968,655	I_kwDOArmXAs53NRmP	62,419	StatefulRandomBinomial not able to run using GPU	{ "login": "tsbrett", "id": 12188485, "node_id": "MDQ6VXNlcjEyMTg4NDg1", "avatar_url": "https://avatars.githubusercontent.com/u/12188485?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tsbrett", "html_url": "https://github.com/tsbrett", "followers_url": "https://api.github.com/users/tsbrett/followers", "following_url": "https://api.github.com/users/tsbrett/following{/other_user}", "gists_url": "https://api.github.com/users/tsbrett/gists{/gist_id}", "starred_url": "https://api.github.com/users/tsbrett/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tsbrett/subscriptions", "organizations_url": "https://api.github.com/users/tsbrett/orgs", "repos_url": "https://api.github.com/users/tsbrett/repos", "events_url": "https://api.github.com/users/tsbrett/events{/privacy}", "received_events_url": "https://api.github.com/users/tsbrett/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473173272, "node_id": "MDU6TGFiZWw0NzMxNzMyNzI=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:feature", "name": "type:feature", "color": "159b2e", "default": false, "description": "Feature requests" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 1097547538, "node_id": "MDU6TGFiZWwxMDk3NTQ3NTM4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:gpu", "name": "comp:gpu", "color": "0052cc", "default": false, "description": "GPU related issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@tsbrett,\r\nApologies for the delay. I tried to execute the mentioned code on tensorflow v2.15 and it was executed without any issue/error. Kindly find the gist of it [here](https://colab.research.google.com/gist/tilakrayal/d9e4252c72967ab98e6c0f448ec0e236/untitled1535.ipynb) and confirm whether we are missing anything in this. Thank you!", "@tilakrayal Thanks for your response. The gist you shared contains everything to reproduce the problem, and, as you say, looks to be working correctly. This suggests is issue is with running the code on an Apple M series. I'm using an Apple M3 pro chip with the package tensorflow-metal v1.1.0 installed. \r\n\r\nI've confirmed everything is correctly installed and tensorflow is able to use the GPU device, e.g. for generating normally distributed random numbers (see the above code example).\r\n\r\n", "@sachinprasadhs,\r\nI tried to execute the code on tensorflow v2.15 with GPU, the code was executed with the user mentioned error. But where as on CPU it was executed without any issue/error. Kindly find the gist of it [here](https://colab.research.google.com/gist/tilakrayal/ebeb4118d235ee45c4faa778f8e9c1a2/untitled1539.ipynb).", "Hi,\r\n\r\nAs the error suggests, binomial is not registered on `GPU` kernels for any of the `dtype`. Have changed the issue to feature request. Will hear it from the team if they have any plan to implement this. Thanks!" ]	2023-11-17T21:37:51	2023-11-29T23:57:20	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version v2.15.0-rc1-8-g6887368d6d4 2.15.0 ### Custom code No ### OS platform and distribution MacOS Sonoma 14.1.1 ### Mobile device _No response_ ### Python version 3.11.6 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I'm trying to generate binomial random numbers using the GPU on my macbook with tensorflow-metal installed. Everything works fine using the CPU. In addition, generating i) normally distributed random numbers and ii) stateless binomially distributed random numbers both work fine using the GPU. I did a search and couldn't find any mention of issues with random number generation using tensorflow-metal. The fact normal random numbers and the stateless generator work fine suggests this issue could be due to a bug/oversight. If the reason is instead that the device isn't supported for StatefulRandomBinomial, I'm happy for this issue to be changed to a feature request. The minimal example to reproduce the bug is taken from the documentation, https://www.tensorflow.org/api_docs/python/tf/random/Generator I'm happy to update this issue with any other information you need to reproduce it. Unfortunately I can't test whether this issue is limited to macOS or is more general. Many thanks! ### Standalone code to reproduce the issue ```shell import tensorflow as tf # Example from tensorflow documentation. # Changing the device to '/cpu:0' results in the random numbers being generated correctly with tf.device('/gpu:0'): counts = [10., 20.] # Probability of success. probs = [0.8] rng = tf.random.Generator.from_seed(seed=234) binomial_samples = rng.binomial(shape=[2], counts=counts, probs=probs) # Generation of normally distributed random numbers using the GPU appears to work fine... with tf.device('/gpu:0'): rng = tf.random.Generator.from_seed(seed=234) normal_samples = rng.normal(shape=[1], mean=[0.1], stddev=10) # ... as does using the stateless generator with tf.device('/gpu:0'): counts = [10., 20.] # Probability of success. probs = [0.8] binomial_samples = tf.random.stateless_binomial(shape=[2], seed=[123, 456], counts=counts, probs=probs) ``` ### Relevant log output ```shell 2023-11-17 13:11:45.268096: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M3 Pro 2023-11-17 13:11:45.268120: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 18.00 GB 2023-11-17 13:11:45.268127: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 6.00 GB 2023-11-17 13:11:45.268154: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:306] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support. 2023-11-17 13:11:45.268169: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:272] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>) Traceback (most recent call last): File "~/example.py", line 11, in <module> binomial_samples = rng.binomial(shape=[2], counts=counts, probs=probs) ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "~/filtered_coffee/venv/lib/python3.11/site-packages/tensorflow/python/ops/stateful_random_ops.py", line 870, in binomial return gen_stateful_random_ops.stateful_random_binomial( ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ File "~/filtered_coffee/venv/lib/python3.11/site-packages/tensorflow/python/ops/gen_stateful_random_ops.py", line 241, in stateful_random_binomial _ops.raise_from_not_ok_status(e, name) File "~/filtered_coffee/venv/lib/python3.11/site-packages/tensorflow/python/framework/ops.py", line 5883, in raise_from_not_ok_status raise core._status_to_exception(e) from None # pylint: disable=protected-access ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ tensorflow.python.framework.errors_impl.InvalidArgumentError: Cannot assign a device for operation StatefulRandomBinomial: Could not satisfy explicit device specification '/job:localhost/replica:0/task:0/device:GPU:0' because no supported kernel for GPU devices is available. Colocation Debug Info: Colocation group had the following types and supported devices: Root Member(assigned_device_name_index_=1 requested_device_name_='/job:localhost/replica:0/task:0/device:GPU:0' assigned_device_name_='/job:localhost/replica:0/task:0/device:GPU:0' resource_device_name_='/job:localhost/replica:0/task:0/device:GPU:0' supported_device_types_=[CPU] possible_devices_=[] StatefulRandomBinomial: CPU _Arg: GPU CPU Colocation members, user-requested devices, and framework assigned devices, if any: resource (_Arg) framework assigned device=/job:localhost/replica:0/task:0/device:GPU:0 StatefulRandomBinomial (StatefulRandomBinomial) /job:localhost/replica:0/task:0/device:GPU:0 Op: StatefulRandomBinomial Node attrs: dtype=DT_INT32, S=DT_INT32, T=DT_FLOAT Registered kernels: device='CPU'; dtype in [DT_HALF]; T in [DT_HALF] device='CPU'; dtype in [DT_HALF]; T in [DT_FLOAT] device='CPU'; dtype in [DT_HALF]; T in [DT_DOUBLE] device='CPU'; dtype in [DT_FLOAT]; T in [DT_HALF] device='CPU'; dtype in [DT_FLOAT]; T in [DT_FLOAT] device='CPU'; dtype in [DT_FLOAT]; T in [DT_DOUBLE] device='CPU'; dtype in [DT_DOUBLE]; T in [DT_HALF] device='CPU'; dtype in [DT_DOUBLE]; T in [DT_FLOAT] device='CPU'; dtype in [DT_DOUBLE]; T in [DT_DOUBLE] device='CPU'; dtype in [DT_INT32]; T in [DT_HALF] device='CPU'; dtype in [DT_INT32]; T in [DT_FLOAT] device='CPU'; dtype in [DT_INT32]; T in [DT_DOUBLE] device='CPU'; dtype in [DT_INT64]; T in [DT_HALF] device='CPU'; dtype in [DT_INT64]; T in [DT_FLOAT] device='CPU'; dtype in [DT_INT64]; T in [DT_DOUBLE] [[{{node StatefulRandomBinomial}}]] [Op:StatefulRandomBinomial] name: ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62419/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62419/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62418	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62418/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62418/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62418/events	https://github.com/tensorflow/tensorflow/issues/62418	1,999,845,934	I_kwDOArmXAs53Mzou	62,418	MLIR: TF-OPT build fails for tensorflow 2.10, 2.14	{ "login": "monowaranjum", "id": 19803082, "node_id": "MDQ6VXNlcjE5ODAzMDgy", "avatar_url": "https://avatars.githubusercontent.com/u/19803082?v=4", "gravatar_id": "", "url": "https://api.github.com/users/monowaranjum", "html_url": "https://github.com/monowaranjum", "followers_url": "https://api.github.com/users/monowaranjum/followers", "following_url": "https://api.github.com/users/monowaranjum/following{/other_user}", "gists_url": "https://api.github.com/users/monowaranjum/gists{/gist_id}", "starred_url": "https://api.github.com/users/monowaranjum/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/monowaranjum/subscriptions", "organizations_url": "https://api.github.com/users/monowaranjum/orgs", "repos_url": "https://api.github.com/users/monowaranjum/repos", "events_url": "https://api.github.com/users/monowaranjum/events{/privacy}", "received_events_url": "https://api.github.com/users/monowaranjum/received_events", "type": "User", "site_admin": false }	[ { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 1205615612, "node_id": "MDU6TGFiZWwxMjA1NjE1NjEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:%20ubuntu/linux", "name": "subtype: ubuntu/linux", "color": "b619ea", "default": false, "description": "Ubuntu/Linux Build/Installation Issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @monowaranjum ,\r\n\r\nCould you please check these [instructions](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/compiler/mlir/README.md) to build with MLIR.\r\n\r\nYou need to add a BUILD.bazel file in the LLVM directory like below which seems causing the issue.\r\n\r\n```\r\nLLVM_SRC=... # this the path to the LLVM local source directory you intend to use.\r\ntouch ${LLVM_SRC}/BUILD.bazel ${LLVM_SRC}/WORKSPACE\r\n```\r\n\r\n`bazel build --override_repository=\"llvm-raw=${LLVM_SRC}\" -c opt tensorflow/compiler/mlir:tf-opt`", "Hi, I tried your recommended solution. I could not get it to work. It only worked when I removed the bazel overlay. The command that worked for me was: ```bazel build -c opt <build target>```. ", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62418\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62418\">No</a>\n" ]	2023-11-17T20:06:43	2023-11-20T21:02:39	2023-11-20T21:02:36	NONE	null	null	null	### Issue type Build/Install ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version 2.10, 2.14 ### Custom code Yes ### OS platform and distribution Ubuntu 22.04 ### Mobile device _No response_ ### Python version 3.10 ### Bazel version 5.1.1 ### GCC/compiler version gcc-11, clang-14 ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? tf-opt build fails. It fails to download some of the files and throws java.io.FileNotFoundException and then after a while it just shows the build failed error. I am using bazelisk for bazel version management and the version being used for tf branch rc2.10 is 5.1.1. Python version is 3.10 . ### Standalone code to reproduce the issue ```shell LLVM_SRC=... # this the path to the LLVM local source directory you intend to use. LLVM_BAZEL_OVERLAY=${LLVM_SRC}/bazel # Note: this can be anywhere mkdir -p ${LLVM_BAZEL_OVERLAY} # This will symlink your LLVM sources with the BUILD files to be usable by Bazel. python ${LLVM_SRC}/utils/bazel/overlay_directories.py \ --src ${LLVM_SRC} \ --overlay ${LLVM_SRC}/utils/bazel/llvm-project-overlay/ \ --target ${LLVM_BAZEL_OVERLAY} touch ${LLVM_BAZEL_OVERLAY}/BUILD.bazel ${LLVM_BAZEL_OVERLAY}/WORKSPACE # The complete list is "AArch64", "AMDGPU", "ARM", "NVPTX", "PowerPC", "RISCV", "SystemZ", "X86" echo 'llvm_targets = ["X86"]' > ${LLVM_BAZEL_OVERLAY}/llvm/targets.bzl ``` Then I run the bazel command: ``` bazel build --override_repository=llvm-project=$LLVM_BAZEL_OVERLAY \ -c opt tensorflow/compiler/mlir:tf-opt ``` ``` ### Relevant log output ```shell Expected the build process to complete and build tf-opt target. Here is the error log: bazel build --override_repository=llvm-project=$LLVM_BAZEL_OVERLAY \ -c opt tensorflow/compiler/mlir:tf-opt 2023/11/17 14:51:11 Downloading https://releases.bazel.build/5.1.1/release/bazel-5.1.1-linux-x86_64... Downloading: 49 MB out of 49 MB (100%) Extracting Bazel installation... Starting local Bazel server and connecting to it... INFO: Options provided by the client: Inherited 'common' options: --isatty=1 --terminal_columns=133 INFO: Reading rc options for 'build' from /home/rashik/Documents/onnx-mlir/virtual_env/tf2onnx-env/tensorflow/.bazelrc: Inherited 'common' options: --experimental_repo_remote_exec INFO: Reading rc options for 'build' from /home/rashik/Documents/onnx-mlir/virtual_env/tf2onnx-env/tensorflow/.bazelrc: 'build' options: --define framework_shared_object=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=standalone -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --enable_platform_specific_config --define=with_xla_support=true --config=short_logs --config=v2 --define=no_aws_support=true --define=no_hdfs_support=true --experimental_cc_shared_library --experimental_link_static_libraries_once=false --deleted_packages=tensorflow/compiler/mlir/tfrt,tensorflow/compiler/mlir/tfrt/benchmarks,tensorflow/compiler/mlir/tfrt/jit/python_binding,tensorflow/compiler/mlir/tfrt/jit/transforms,tensorflow/compiler/mlir/tfrt/python_tests,tensorflow/compiler/mlir/tfrt/tests,tensorflow/compiler/mlir/tfrt/tests/ir,tensorflow/compiler/mlir/tfrt/tests/analysis,tensorflow/compiler/mlir/tfrt/tests/jit,tensorflow/compiler/mlir/tfrt/tests/lhlo_to_tfrt,tensorflow/compiler/mlir/tfrt/tests/lhlo_to_jitrt,tensorflow/compiler/mlir/tfrt/tests/tf_to_corert,tensorflow/compiler/mlir/tfrt/tests/tf_to_tfrt_data,tensorflow/compiler/mlir/tfrt/tests/saved_model,tensorflow/compiler/mlir/tfrt/transforms/lhlo_gpu_to_tfrt_gpu,tensorflow/core/runtime_fallback,tensorflow/core/runtime_fallback/conversion,tensorflow/core/runtime_fallback/kernel,tensorflow/core/runtime_fallback/opdefs,tensorflow/core/runtime_fallback/runtime,tensorflow/core/runtime_fallback/util,tensorflow/core/tfrt/common,tensorflow/core/tfrt/eager,tensorflow/core/tfrt/eager/backends/cpu,tensorflow/core/tfrt/eager/backends/gpu,tensorflow/core/tfrt/eager/core_runtime,tensorflow/core/tfrt/eager/cpp_tests/core_runtime,tensorflow/core/tfrt/gpu,tensorflow/core/tfrt/run_handler_thread_pool,tensorflow/core/tfrt/runtime,tensorflow/core/tfrt/saved_model,tensorflow/core/tfrt/graph_executor,tensorflow/core/tfrt/saved_model/tests,tensorflow/core/tfrt/tpu,tensorflow/core/tfrt/utils INFO: Found applicable config definition build:short_logs in file /home/rashik/Documents/onnx-mlir/virtual_env/tf2onnx-env/tensorflow/.bazelrc: --output_filter=DONT_MATCH_ANYTHING INFO: Found applicable config definition build:v2 in file /home/rashik/Documents/onnx-mlir/virtual_env/tf2onnx-env/tensorflow/.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1 INFO: Found applicable config definition build:linux in file /home/rashik/Documents/onnx-mlir/virtual_env/tf2onnx-env/tensorflow/.bazelrc: --copt=-w --host_copt=-w --define=PREFIX=/usr --define=LIBDIR=$(PREFIX)/lib --define=INCLUDEDIR=$(PREFIX)/include --define=PROTOBUF_INCLUDE_PATH=$(PREFIX)/include --cxxopt=-std=c++17 --host_cxxopt=-std=c++17 --config=dynamic_kernels --distinct_host_configuration=false --experimental_guard_against_concurrent_changes INFO: Found applicable config definition build:dynamic_kernels in file /home/rashik/Documents/onnx-mlir/virtual_env/tf2onnx-env/tensorflow/.bazelrc: --define=dynamic_loaded_kernels=true --copt=-DAUTOLOAD_DYNAMIC_KERNELS WARNING: Download from https://storage.googleapis.com/mirror.tensorflow.org/github.com/tensorflow/runtime/archive/6ca793b5d862ef6c50f242d77a811f06cce9b60a.tar.gz failed: class java.io.FileNotFoundException GET returned 404 Not Found WARNING: Download from https://storage.googleapis.com/mirror.tensorflow.org/github.com/llvm/llvm-project/archive/0538e5431afdb1fa05bdcedf70ee502ccfcd112a.tar.gz failed: class java.io.FileNotFoundException GET returned 404 Not Found ERROR: Skipping 'tensorflow/compiler/mlir:tf-opt': no such package 'tensorflow/compiler/mlir/tensorflow/compiler/mlir': BUILD file not found in any of the following directories. Add a BUILD file to a directory to mark it as a package. - /home/rashik/Documents/onnx-mlir/virtual_env/tf2onnx-env/tensorflow/tensorflow/compiler/mlir/tensorflow/compiler/mlir WARNING: Target pattern parsing failed. ERROR: no such package 'tensorflow/compiler/mlir/tensorflow/compiler/mlir': BUILD file not found in any of the following directories. Add a BUILD file to a directory to mark it as a package. - /home/rashik/Documents/onnx-mlir/virtual_env/tf2onnx-env/tensorflow/tensorflow/compiler/mlir/tensorflow/compiler/mlir INFO: Elapsed time: 37.349s INFO: 0 processes. FAILED: Build did NOT complete successfully (0 packages loaded) ``` ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62418/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62418/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62417	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62417/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62417/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62417/events	https://github.com/tensorflow/tensorflow/pull/62417	1,999,648,951	PR_kwDOArmXAs5fx01o	62,417	Changes to add bfloat16 support for image rotation	{ "login": "jojivk73", "id": 14943401, "node_id": "MDQ6VXNlcjE0OTQzNDAx", "avatar_url": "https://avatars.githubusercontent.com/u/14943401?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jojivk73", "html_url": "https://github.com/jojivk73", "followers_url": "https://api.github.com/users/jojivk73/followers", "following_url": "https://api.github.com/users/jojivk73/following{/other_user}", "gists_url": "https://api.github.com/users/jojivk73/gists{/gist_id}", "starred_url": "https://api.github.com/users/jojivk73/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/jojivk73/subscriptions", "organizations_url": "https://api.github.com/users/jojivk73/orgs", "repos_url": "https://api.github.com/users/jojivk73/repos", "events_url": "https://api.github.com/users/jojivk73/events{/privacy}", "received_events_url": "https://api.github.com/users/jojivk73/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169364458, "node_id": "MDU6TGFiZWwxMTY5MzY0NDU4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:S", "name": "size:S", "color": "adafea", "default": false, "description": "CL Change Size: Small" }, { "id": 1478826728, "node_id": "MDU6TGFiZWwxNDc4ODI2NzI4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:core", "name": "comp:core", "color": "024391", "default": false, "description": "issues related to core part of tensorflow" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @jojivk73 Can you please check @cantonios's comments and keep us posted ? Thank you!", "@gbaned @cantonios. Sorry for the delay. I updated the name for testcase.\r\nThanks" ]	2023-11-17T17:52:46	2023-12-16T18:53:37	2023-12-16T18:53:36	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62417", "html_url": "https://github.com/tensorflow/tensorflow/pull/62417", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62417.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62417.patch", "merged_at": "2023-12-16T18:53:36" }	These are changes to add bfloat16 for image rotation. With mixed bfloat16 precision, models that use image transformation would perform much better with image rotation in bfloat16.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62417/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62417/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62416	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62416/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62416/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62416/events	https://github.com/tensorflow/tensorflow/issues/62416	1,999,556,811	I_kwDOArmXAs53LtDL	62,416	TF 2.15 fails to build on Ubuntu 23.10	{ "login": "JohnRTitor", "id": 50095635, "node_id": "MDQ6VXNlcjUwMDk1NjM1", "avatar_url": "https://avatars.githubusercontent.com/u/50095635?v=4", "gravatar_id": "", "url": "https://api.github.com/users/JohnRTitor", "html_url": "https://github.com/JohnRTitor", "followers_url": "https://api.github.com/users/JohnRTitor/followers", "following_url": "https://api.github.com/users/JohnRTitor/following{/other_user}", "gists_url": "https://api.github.com/users/JohnRTitor/gists{/gist_id}", "starred_url": "https://api.github.com/users/JohnRTitor/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/JohnRTitor/subscriptions", "organizations_url": "https://api.github.com/users/JohnRTitor/orgs", "repos_url": "https://api.github.com/users/JohnRTitor/repos", "events_url": "https://api.github.com/users/JohnRTitor/events{/privacy}", "received_events_url": "https://api.github.com/users/JohnRTitor/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1205615612, "node_id": "MDU6TGFiZWwxMjA1NjE1NjEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:%20ubuntu/linux", "name": "subtype: ubuntu/linux", "color": "b619ea", "default": false, "description": "Ubuntu/Linux Build/Installation Issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	closed	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "I faced the same issue and managed to work around it by adding the ``--copt=-Wno-gnu-offsetof-extensions`` flag.", "Added the --copt=-Wno-gnu-offsetof-extensions flag.\r\nNow it tries to compile all to the end, instead of failing midway. And then it fails at the last moment, just at the last step.\r\n\r\n```\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/python/platform/BUILD:388:17: Linking tensorflow/python/platform/_pywrap_tf2.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/python/platform:_pywrap_tf2.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/python/platform/_pywrap_tf2.so-2.params\r\nld.lld: error: version script assignment of 'global' to symbol 'init_pywrap_tf2' failed: symbol not defined\r\nld.lld: error: version script assignment of 'global' to symbol 'init__pywrap_tf2' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nTarget //tensorflow/tools/pip_package:build_pip_package failed to build\r\nUse --verbose_failures to see the command lines of failed build steps.\r\nINFO: Elapsed time: 3962.530s, Critical Path: 167.04s\r\nINFO: 11472 processes: 14 internal, 11458 local.\r\nFAILED: Build did NOT complete successfully\r\n\r\n```", "Running with the -k flag in bazel build:\r\n\r\n```\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/python/platform/BUILD:388:17: Linking tensorflow/python/platform/_pywrap_tf2.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/python/platform:_pywrap_tf2.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/python/platform/_pywrap_tf2.so-2.params\r\nld.lld: error: version script assignment of 'global' to symbol 'init_pywrap_tf2' failed: symbol not defined\r\nld.lld: error: version script assignment of 'global' to symbol 'init__pywrap_tf2' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/compiler/tf2tensorrt/BUILD:1012:17: Linking tensorflow/compiler/tf2tensorrt/_pywrap_py_utils.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/compiler/tf2tensorrt:_pywrap_py_utils.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/compiler/tf2tensorrt/_pywrap_py_utils.so-2.params\r\nld.lld: error: version script assignment of 'global' to symbol 'init_pywrap_py_utils' failed: symbol not defined\r\nld.lld: error: version script assignment of 'global' to symbol 'init__pywrap_py_utils' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/python/autograph/impl/testing/BUILD:9:17: Linking tensorflow/python/autograph/impl/testing/pybind_for_testing.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/python/autograph/impl/testing:pybind_for_testing.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/python/autograph/impl/testing/pybind_for_testing.so-2.params\r\nld.lld: error: version script assignment of 'global' to symbol 'initpybind_for_testing' failed: symbol not defined\r\nld.lld: error: version script assignment of 'global' to symbol 'init_pybind_for_testing' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/lite/python/metrics/BUILD:35:17: Linking tensorflow/lite/python/metrics/_pywrap_tensorflow_lite_metrics_wrapper.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/lite/python/metrics:_pywrap_tensorflow_lite_metrics_wrapper.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/lite/python/metrics/_pywrap_tensorflow_lite_metrics_wrapper.so-2.params\r\nld.lld: error: version script assignment of 'global' to symbol 'init_pywrap_tensorflow_lite_metrics_wrapper' failed: symbol not defined\r\nld.lld: error: version script assignment of 'global' to symbol 'init__pywrap_tensorflow_lite_metrics_wrapper' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/lite/python/interpreter_wrapper/BUILD:79:17: Linking tensorflow/lite/python/interpreter_wrapper/_pywrap_tensorflow_interpreter_wrapper.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/lite/python/interpreter_wrapper:_pywrap_tensorflow_interpreter_wrapper.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/lite/python/interpreter_wrapper/_pywrap_tensorflow_interpreter_wrapper.so-2.params\r\nld.lld: error: version script assignment of 'global' to symbol 'init_pywrap_tensorflow_interpreter_wrapper' failed: symbol not defined\r\nld.lld: error: version script assignment of 'global' to symbol 'init__pywrap_tensorflow_interpreter_wrapper' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/compiler/mlir/stablehlo/BUILD:20:21: Linking tensorflow/compiler/mlir/stablehlo/stablehlo_extension.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/compiler/mlir/stablehlo:stablehlo_extension.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/compiler/mlir/stablehlo/stablehlo_extension.so-2.params\r\nld.lld: error: version script assignment of 'global' to symbol 'initstablehlo_extension' failed: symbol not defined\r\nld.lld: error: version script assignment of 'global' to symbol 'init_stablehlo_extension' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/lite/python/optimize/BUILD:38:17: Linking tensorflow/lite/python/optimize/_pywrap_tensorflow_lite_calibration_wrapper.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/lite/python/optimize:_pywrap_tensorflow_lite_calibration_wrapper.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/lite/python/optimize/_pywrap_tensorflow_lite_calibration_wrapper.so-2.params\r\nld.lld: error: version script assignment of 'global' to symbol 'init_pywrap_tensorflow_lite_calibration_wrapper' failed: symbol not defined\r\nld.lld: error: version script assignment of 'global' to symbol 'init__pywrap_tensorflow_lite_calibration_wrapper' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/lite/python/analyzer_wrapper/BUILD:9:17: Linking tensorflow/lite/python/analyzer_wrapper/_pywrap_analyzer_wrapper.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/lite/python/analyzer_wrapper:_pywrap_analyzer_wrapper.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/lite/python/analyzer_wrapper/_pywrap_analyzer_wrapper.so-2.params\r\nld.lld: error: version script assignment of 'global' to symbol 'init_pywrap_analyzer_wrapper' failed: symbol not defined\r\nld.lld: error: version script assignment of 'global' to symbol 'init__pywrap_analyzer_wrapper' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nERROR: /home/masum/Desktop/Dev-Enviornment/tensorflow/tensorflow/python/BUILD:646:24: Linking tensorflow/python/_pywrap_tensorflow_internal.so failed: (Exit 1): clang failed: error executing command (from target //tensorflow/python:_pywrap_tensorflow_internal.so) /usr/lib/llvm-17/bin/clang @bazel-out/k8-opt/bin/tensorflow/python/_pywrap_tensorflow_internal.so-2.params\r\nld.lld: error: version script assignment of 'tensorflow' to symbol 'init_pywrap_tensorflow_internal' failed: symbol not defined\r\nld.lld: error: version script assignment of 'tensorflow' to symbol '_init_pywrap_tensorflow_internal' failed: symbol not defined\r\nclang: error: linker command failed with exit code 1 (use -v to see invocation)\r\nTarget //tensorflow/tools/pip_package:build_pip_package failed to build\r\nUse --verbose_failures to see the command lines of failed build steps.\r\nINFO: Elapsed time: 131.887s, Critical Path: 130.84s\r\nINFO: 555 processes: 76 internal, 479 local.\r\nFAILED: Build did NOT complete successfully\r\n\r\n```", "Quick note since this is the first google result for this error -- this happens when running with clang 17 or 18, so if you're just trying to build tensorflow then follow the [documentation](https://www.tensorflow.org/install/source#install_clang_recommended_linux_only) and use LLVM 16 / clang 16.", "Does using clang 16 builds it completely without error? In my case, both were installed. But the configure script preferred clang-17 for some reason for r2.15. For 2.14 and below it prefers clang-14.\r\n\r\n\r\nOn another note, I managed to completely workaround this issue by using a container to build tensorflow instead of using packages and libraries from my host system.\r\nI used Distrobox with Debian-Testing image, installed the necessary packages, used a Python virtual enviornment. And just followed the instructions. Built successfully without any library errors. ", "@azza-bazoo the documentation now says use Clang-17....", "@JohnRTitor ,\r\nPlease use the below tested build configurations\r\n\r\n\r\nVersion \| Python version \| Compiler \| Build tools\r\n-- \| -- \| -- \| --\r\ntensorflow-2.16.1 \| 3.9-3.12 \| Clang 17.0.6 \| Bazel 6.5.0\r\n\r\n", "I do not run Ubuntu 23.10 on my main system anymore and instead have switched to NixOS. Can other participants like @azza-bazoo @AIWintermuteAI test this with the above build tools on Ubuntu 23.10?\r\n\r\nNote: to get a specific version of Clang you can use https://apt.llvm.org/ and as for Bazel, just use bazelisk. ", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62416\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62416\">No</a>\n", "It seems like unnecessary symbol ignores causing issues. At Solus this is how we fixed it with Clang 18: https://github.com/getsolus/packages/blob/dfc56ba57a8af8233a635e309b499ff5d27992f4/packages/t/tensorflow/files/fix-clang-18.diff" ]	2023-11-17T16:54:13	2024-06-05T14:35:14	2024-04-04T01:48:01	NONE	null	null	null	### Issue type Build/Install ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.15 ### Custom code No ### OS platform and distribution Ubuntu 23.10 ### Mobile device _No response_ ### Python version 3.11.5 Anaconda ### Bazel version 6.1.0 ### GCC/compiler version Clang 17.0.4 ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Hi, I was trying to build the release 2.15 myself with the flag march=native. But unfortunately it fails when just building the package builder program using bazel. I use a Ryzen 5 7600 CPU with no dedicated graphics (has integrated Radeon graphics). So I was trying to build a CPU only version. (no ROCM, no CUDA. The compiler used was Clang 17.0.4. ### Standalone code to reproduce the issue ```shell git checkout r2.15 export TF_PYTHON_VERSION=3.11 ./configure # Use -march=native as compiler optimization flag bazel build --config=opt //tensorflow/tools/pip_package:build_pip_package # It fails building here ``` ### Relevant log output ```shell ERROR: /home/masum/.cache/bazel/_bazel_masum/353a8711dbc9282a1a4f3a8a30dc5597/external/upb/BUILD:57:11: Compiling upb/upb.c failed: (Exit 1): clang failed: error executing command (from target @upb//:upb) /usr/lib/llvm-17/bin/clang -U_FORTIFY_SOURCE -fstack-protector -Wall -Wthread-safety -Wself-assign -Wunused-but-set-parameter -Wno-free-nonheap-object -fcolor-diagnostics -fno-omit-frame-pointer -g0 ... (remaining 41 arguments skipped) external/upb/upb/upb.c:192:10: error: defining a type within 'offsetof' is a Clang extension [-Werror,-Wgnu-offsetof-extensions] 192 \| n &= ~(upb_alignof(upb_arena) - 1); \| ^~~~~~~~~~~~~~~~~~~~~~ external/upb/upb/upb.c:183:37: note: expanded from macro 'upb_alignof' 183 \| #define upb_alignof(type) offsetof (struct { char c; type member; }, member) \| ^~~~~~ /usr/lib/llvm-17/lib/clang/17/include/stddef.h:116:43: note: expanded from macro 'offsetof' 116 \| #define offsetof(t, d) __builtin_offsetof(t, d) \| ^ 1 error generated. Target //tensorflow/tools/pip_package:build_pip_package failed to build Use --verbose_failures to see the command lines of failed build steps. INFO: Elapsed time: 10.105s, Critical Path: 2.83s INFO: 23 processes: 13 internal, 10 local. FAILED: Build did NOT complete successfully ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62416/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 1 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62416/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62415	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62415/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62415/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62415/events	https://github.com/tensorflow/tensorflow/issues/62415	1,999,050,283	I_kwDOArmXAs53JxYr	62,415	WARNING:tensorflow	{ "login": "mominurr", "id": 105736335, "node_id": "U_kgDOBk1ojw", "avatar_url": "https://avatars.githubusercontent.com/u/105736335?v=4", "gravatar_id": "", "url": "https://api.github.com/users/mominurr", "html_url": "https://github.com/mominurr", "followers_url": "https://api.github.com/users/mominurr/followers", "following_url": "https://api.github.com/users/mominurr/following{/other_user}", "gists_url": "https://api.github.com/users/mominurr/gists{/gist_id}", "starred_url": "https://api.github.com/users/mominurr/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/mominurr/subscriptions", "organizations_url": "https://api.github.com/users/mominurr/orgs", "repos_url": "https://api.github.com/users/mominurr/repos", "events_url": "https://api.github.com/users/mominurr/events{/privacy}", "received_events_url": "https://api.github.com/users/mominurr/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473184161, "node_id": "MDU6TGFiZWw0NzMxODQxNjE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:support", "name": "type:support", "color": "159b2e", "default": false, "description": "Support issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1097545817, "node_id": "MDU6TGFiZWwxMDk3NTQ1ODE3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:apis", "name": "comp:apis", "color": "0052cc", "default": false, "description": "Highlevel API related issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	closed	false	{ "login": "Venkat6871", "id": 147127861, "node_id": "U_kgDOCMT-NQ", "avatar_url": "https://avatars.githubusercontent.com/u/147127861?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Venkat6871", "html_url": "https://github.com/Venkat6871", "followers_url": "https://api.github.com/users/Venkat6871/followers", "following_url": "https://api.github.com/users/Venkat6871/following{/other_user}", "gists_url": "https://api.github.com/users/Venkat6871/gists{/gist_id}", "starred_url": "https://api.github.com/users/Venkat6871/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Venkat6871/subscriptions", "organizations_url": "https://api.github.com/users/Venkat6871/orgs", "repos_url": "https://api.github.com/users/Venkat6871/repos", "events_url": "https://api.github.com/users/Venkat6871/events{/privacy}", "received_events_url": "https://api.github.com/users/Venkat6871/received_events", "type": "User", "site_admin": false }	[ { "login": "Venkat6871", "id": 147127861, "node_id": "U_kgDOCMT-NQ", "avatar_url": "https://avatars.githubusercontent.com/u/147127861?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Venkat6871", "html_url": "https://github.com/Venkat6871", "followers_url": "https://api.github.com/users/Venkat6871/followers", "following_url": "https://api.github.com/users/Venkat6871/following{/other_user}", "gists_url": "https://api.github.com/users/Venkat6871/gists{/gist_id}", "starred_url": "https://api.github.com/users/Venkat6871/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Venkat6871/subscriptions", "organizations_url": "https://api.github.com/users/Venkat6871/orgs", "repos_url": "https://api.github.com/users/Venkat6871/repos", "events_url": "https://api.github.com/users/Venkat6871/events{/privacy}", "received_events_url": "https://api.github.com/users/Venkat6871/received_events", "type": "User", "site_admin": false } ]	null	[ "The warning is just telling you something used during the import is deprecated. Can be safely ignored", "Hi @mominurr ,\r\n\r\nCould you please try again? I tried with TF2.15 and I cannot reproduce the error. Please check the [gist](https://colab.research.google.com/gist/Venkat6871/4c04c8ff77c5772c556693cefa913f48/62415_2-15-v.ipynb) here. \r\n\r\nThank you!", "@mominurr This error is only showing to using older version just update the tensorflow to latest and this will be resolved. This issue will be resolved \r\n\r\n![image](https://github.com/tensorflow/tensorflow/assets/67970591/57385de8-ae15-4647-ad06-88d440d646db)\r\n", "I'm getting the same warning with tensorflow 2.15.0. Was installed with _pip install tensorflow==2.15._\r\n\r\n```\r\nWARNING:tensorflow:From [c:\\Users\\xxxxx\\miniconda3\\envs\\xxxxx\\lib\\site-packages\\keras\\src\\losses.py:2976](file:///C:/Users/xxxxx/miniconda3/envs/xxxxx/lib/site-packages/keras/src/losses.py:2976): The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\r\n```\r\n\r\nThe warning is being generated from keras.losses.py, Line 2976:\r\n\r\n```\r\nLABEL_DTYPES_FOR_LOSSES = {\r\n tf.compat.v1.losses.sparse_softmax_cross_entropy: \"int32\",\r\n sparse_categorical_crossentropy: \"int32\",\r\n}\r\n```\r\n\r\n(so this warning doesn't make any sense because the code uses `tf.compat.v1.losses.sparse_softmax_cross_entropy`)", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "> ### Issue type\r\n> Others\r\n> \r\n> ### Have you reproduced the bug with TensorFlow Nightly?\r\n> Yes\r\n> \r\n> ### Source\r\n> source\r\n> \r\n> ### TensorFlow version\r\n> 2.15.0\r\n> \r\n> ### Custom code\r\n> Yes\r\n> \r\n> ### OS platform and distribution\r\n> windows\r\n> \r\n> ### Mobile device\r\n> _No response_\r\n> \r\n> ### Python version\r\n> 3.10.10\r\n> \r\n> ### Bazel version\r\n> _No response_\r\n> \r\n> ### GCC/compiler version\r\n> _No response_\r\n> \r\n> ### CUDA/cuDNN version\r\n> _No response_\r\n> \r\n> ### GPU model and memory\r\n> _No response_\r\n> \r\n> ### Current behavior?\r\n> import os os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0' import tensorflow as tf\r\n> \r\n> node1 = tf.constant(3, dtype=tf.int64) node2 = tf.constant(4, dtype=tf.int64) node3= tf.add(node1, node2)\r\n> \r\n> # Directly print the result\r\n> print(\"sum of node1 and node2 is:\", node3.numpy())\r\n> \r\n> ### Standalone code to reproduce the issue\r\n> ```shell\r\n> for this code give me this warning. How I can fixed it.\r\n> \r\n> WARNING:tensorflow:From D:\\Learning-Module\\Data science practise\\deap-learning-with-tenserflow\\env\\lib\\site-packages\\keras\\src\\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\r\n> \r\n> 2023-11-17 18:31:23.771915: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\r\n> To enable the following instructions: SSE SSE2 SSE3 SSE4.1 SSE4.2 AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\r\n> ```\r\n> \r\n> ### Relevant log output\r\n> you can go to \\lib\\site-packages\\keras\\src\\losses.py file and write \r\n\r\n import tensorflow.python as tf* \r\n\r\nand the last of file you can modify like this: \r\n\r\n LABEL_DTYPES_FOR_LOSSES = {\r\n tf.losses.sparse_softmax_cross_entropy: \"int32\",\r\n tf.keras.losses.sparse_categorical_crossentropy: \"int32\",\r\n} \r\n\r\n![for_losses](https://github.com/tensorflow/tensorflow/assets/58201720/aa9cda95-51e8-48c4-a998-c53671af726d)\r\n![tf python](https://github.com/tensorflow/tensorflow/assets/58201720/105195fd-546e-4eb2-9969-e2db3346a4a6)\r\n\r\n", "@MaoMakara This solution does remove the console warning! Thank you.", "@MaoMakara thanks for suggesting a temporary solution for local/personal projects. Still looking for an official fix from the Tensorflow/Keras team", "> > ### Issue type\r\n> > Others\r\n> > ### Have you reproduced the bug with TensorFlow Nightly?\r\n> > Yes\r\n> > ### Source\r\n> > source\r\n> > ### TensorFlow version\r\n> > 2.15.0\r\n> > ### Custom code\r\n> > Yes\r\n> > ### OS platform and distribution\r\n> > windows\r\n> > ### Mobile device\r\n> > _No response_\r\n> > ### Python version\r\n> > 3.10.10\r\n> > ### Bazel version\r\n> > _No response_\r\n> > ### GCC/compiler version\r\n> > _No response_\r\n> > ### CUDA/cuDNN version\r\n> > _No response_\r\n> > ### GPU model and memory\r\n> > _No response_\r\n> > ### Current behavior?\r\n> > import os os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0' import tensorflow as tf\r\n> > node1 = tf.constant(3, dtype=tf.int64) node2 = tf.constant(4, dtype=tf.int64) node3= tf.add(node1, node2)\r\n> > # Directly print the result\r\n> > print(\"sum of node1 and node2 is:\", node3.numpy())\r\n> > ### Standalone code to reproduce the issue\r\n> > ```shell\r\n> > for this code give me this warning. How I can fixed it.\r\n> > \r\n> > WARNING:tensorflow:From D:\\Learning-Module\\Data science practise\\deap-learning-with-tenserflow\\env\\lib\\site-packages\\keras\\src\\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\r\n> > \r\n> > 2023-11-17 18:31:23.771915: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\r\n> > To enable the following instructions: SSE SSE2 SSE3 SSE4.1 SSE4.2 AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\r\n> > ```\r\n> > \r\n> > \r\n> > \r\n> > \r\n> > \r\n> > \r\n> > \r\n> > \r\n> > \r\n> > \r\n> > \r\n> > ### Relevant log output\r\n> > you can go to \\lib\\site-packages\\keras\\src\\losses.py file and write\r\n> \r\n> import tensorflow.python as tf\r\n> \r\n> and the last of file you can modify like this:\r\n> \r\n> LABEL_DTYPES_FOR_LOSSES = { tf.losses.sparse_softmax_cross_entropy: \"int32\", tf.keras.losses.sparse_categorical_crossentropy: \"int32\", }\r\n> \r\n> ![for_losses](https://private-user-images.githubusercontent.com/58201720/289242627-aa9cda95-51e8-48c4-a998-c53671af726d.PNG?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.IZ2j_na8iiS02451UsKkM240Y5Extjg91HKARIdnMWA) ![tf python](https://private-user-images.githubusercontent.com/58201720/289242638-105195fd-546e-4eb2-9969-e2db3346a4a6.png?jwt=eyJhbGciOiJIUzI1NiIsInR5cCI6IkpXVCJ9.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.HF9t5slKZ33u_DcNG02SeaqUn4PaslyJtF0r3y2wc3M)\r\n\r\nThanks. the changes directly to the code works. I am \"not\" annoyed now by tensorflow,even though i am not using the library, it fills my terminal with this warnings. ", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "@MaoMakara i got this error when do ur change to my code, AttributeError: module 'tensorflow.python' has no attribute '__internal__'", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62415\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62415\">No</a>\n", "Hi @mominurr ,\r\n\r\nThe warning is just telling users something about the deprecated during the import. Tensorflow will provide such warnings/info logs to the users for their understanding.\r\nIf users use that deprecated apis then also it will raise the warnings.\r\n\r\nThank you!", "The warnings as per are not the issue. We appreciate them. \r\n\r\nThe issue is they filling the terminal up, and we are not able to suppress them somehow after recognising them. \r\n\r\n> Hi @mominurr** ,\r\n> \r\n> The warning is just telling users something about the deprecated during the import. Tensorflow will provide such warnings/info logs to the users for their understanding. If users use that deprecated apis then also it will raise the warnings.\r\n> \r\n> Thank you!\r\n\r\n", "> ### Issue type\r\n> Others\r\n> \r\n> ### Have you reproduced the bug with TensorFlow Nightly?\r\n> Yes\r\n> \r\n> ### Source\r\n> source\r\n> \r\n> ### TensorFlow version\r\n> 2.15.0\r\n> \r\n> ### Custom code\r\n> Yes\r\n> \r\n> ### OS platform and distribution\r\n> windows\r\n> \r\n> ### Mobile device\r\n> _No response_\r\n> \r\n> ### Python version\r\n> 3.10.10\r\n> \r\n> ### Bazel version\r\n> _No response_\r\n> \r\n> ### GCC/compiler version\r\n> _No response_\r\n> \r\n> ### CUDA/cuDNN version\r\n> _No response_\r\n> \r\n> ### GPU model and memory\r\n> _No response_\r\n> \r\n> ### Current behavior?\r\n> import os os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0' import tensorflow as tf\r\n> \r\n> node1 = tf.constant(3, dtype=tf.int64) node2 = tf.constant(4, dtype=tf.int64) node3= tf.add(node1, node2)\r\n> \r\n> # Directly print the result\r\n> print(\"sum of node1 and node2 is:\", node3.numpy())\r\n> \r\n> ### Standalone code to reproduce the issue\r\n> ```shell\r\n> for this code give me this warning. How I can fixed it.\r\n> \r\n> WARNING:tensorflow:From D:\\Learning-Module\\Data science practise\\deap-learning-with-tenserflow\\env\\lib\\site-packages\\keras\\src\\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\r\n> \r\n> 2023-11-17 18:31:23.771915: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\r\n> To enable the following instructions: SSE SSE2 SSE3 SSE4.1 SSE4.2 AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\r\n> ```\r\n> \r\n> ### Relevant log output\r\n> _No response_\r\n\r\nhi, were you able to fix the \"The name tf.losses.sparse_softmax_cross_entropy is deprecated\" ERROR?", "Hi @sujirou I am also getting the same error and the previously mentioned solution does not work since I get the error:\r\n```AttributeError: module 'tensorflow.python' has no attribute '__internal__'```\r\n\r\nI have expanded on my issue here: https://stackoverflow.com/questions/77936905/tensorflow-core-util-port-cc113-onednn-custom-operations-are-on-the-name-tf-l", "Hi all.\r\n@Venkat6871 and @mihaimaruseac , yes, it's just a warning, but still annoying. It's not rooted in the user's code, and therefore, in addition to being annoying, it might be misleading and result in a waste of time. It needs to be fixed!\r\nI'm getting this warning by just simply importing tensorflow and nothing else!\r\n\r\n@MaoMakara Thanks for the quick fix. It worked at first, but as I developed my project, at some point, somehow, it resulted in an error in my code. Changing it back solved the problem, so be aware that the change you are suggesting might raise an error.\r\nA safer solution looks like this:\r\n\r\n# Quick fix\r\nThe warning is printed by [tensorflow/python/util/module_wrapper.py, Line 149-151](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/util/module_wrapper.py#L149-L151). Simply comment these lines and you are free of its warning's pain.", "If it results from simply importing TF, the good fix is to identify the code that causes the warning to occur and fix that. The code in `module_wrapper.py` is there to prevent such cases, supposedly.", "> # Quick fix\r\n> The warning is printed by [tensorflow/python/util/module_wrapper.py, Line 149-151](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/python/util/module_wrapper.py#L149-L151). Simply comment these lines and you are free of its warning's pain.\r\n\r\n* yep. this works fine for me. thank you very much!\r\n\r\n* previously i changed the code in `...\\lib\\site-packages\\keras\\src\\losses.py`, the warning was gone, however, it then caused another error `AttributeError: module 'tensorflow.python' has no attribute '__internal__'` when i ran OpenAI `baselines`. for some reason i had to used older `tensorflow` version, so suppressing the warning function works the best for me. \r\n\r\n* in file `D:\\Users\\*\\miniconda3\\envs\\drlnd_py311\\Lib\\site-packages\\tensorflow\\python\\util\\module_wrapper.py`:\r\n```\r\n def _tfmw_add_deprecation_warning(self, name, attr):\r\n return False ## ---------------------------------- 👈✅ i added this line to bypass the whole function.\r\n \"\"\"Print deprecation warning for attr with given name if necessary.\"\"\"\r\n if (self._tfmw_warning_count < _PER_MODULE_WARNING_LIMIT and\r\n name not in self._tfmw_deprecated_checked):\r\n\r\n self._tfmw_deprecated_checked.add(name)\r\n\r\n if self._tfmw_module_name:\r\n full_name = 'tf.%s.%s' % (self._tfmw_module_name, name)\r\n else:\r\n full_name = 'tf.%s' % name\r\n rename = get_rename_v2(full_name)\r\n if rename and not has_deprecation_decorator(attr):\r\n call_location = _call_location()\r\n # skip locations in Python source\r\n if not call_location.startswith('<'):\r\n logging.warning(\r\n 'From %s: The name %s is deprecated. Please use %s instead.\\n',\r\n _call_location(), full_name, rename)\r\n self._tfmw_warning_count += 1\r\n return True\r\n return False\r\n```\r\n\r\n", "> The warning is just telling you something used during the import is deprecated. Can be safely ignored\r\n\r\nI disagree with this statement.\r\n\r\nIf it is telling me that it is deprecated that means I should change it because at some point in the future deprecated name will be removed and my code will stop working. At least that's how developers use deprecation warnings in other languages.\r\n\r\nAlso, the warning doesn't make sense.\r\n```\r\nWARNING:\r\ntensorflow:\r\nFrom d:\\AI\\AUTOMATIC1111\\venv\\Lib\\site-packages\\keras\\src\\losses.py:2976:\r\nThe name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead.\r\n```\r\nHow I read it, it is telling us to not use `tf.losses.sparse_softmax_cross_entropy` but `tf.compat.v1.losses.sparse_softmax_cross_entropy` instead.\r\n\r\nHowever, from `d:\\AI\\AUTOMATIC1111\\venv\\Lib\\site-packages\\keras\\src\\losses.py:2976`\r\n```\r\nLABEL_DTYPES_FOR_LOSSES = {\r\n tf.compat.v1.losses.sparse_softmax_cross_entropy: \"int32\",\r\n sparse_categorical_crossentropy: \"int32\",\r\n}\r\n```\r\nWe can see that the correct name is already used. So why the warning then? Please clarify?", "Unsure (cannot check now) but is it possible that `tf.compat.v1.losses.sparse_softmax_cross_entropy` just calls the old `tf.losses.sparse_softmax_cross_entropy` and this causes the deprecation warning to trigger?\r\n\r\nThe deprecations are added as decorators, so they trigger at import time when the module is parsed. Maybe that's too eager." ]	2023-11-17T12:46:22	2024-03-07T16:51:47	2024-01-10T01:49:24	NONE	null	null	null	### Issue type Others ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.15.0 ### Custom code Yes ### OS platform and distribution windows ### Mobile device _No response_ ### Python version 3.10.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? import os os.environ['TF_ENABLE_ONEDNN_OPTS'] = '0' import tensorflow as tf node1 = tf.constant(3, dtype=tf.int64) node2 = tf.constant(4, dtype=tf.int64) node3= tf.add(node1, node2) # Directly print the result print("sum of node1 and node2 is:", node3.numpy()) ### Standalone code to reproduce the issue ```shell for this code give me this warning. How I can fixed it. WARNING:tensorflow:From D:\Learning-Module\Data science practise\deap-learning-with-tenserflow\env\lib\site-packages\keras\src\losses.py:2976: The name tf.losses.sparse_softmax_cross_entropy is deprecated. Please use tf.compat.v1.losses.sparse_softmax_cross_entropy instead. 2023-11-17 18:31:23.771915: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: SSE SSE2 SSE3 SSE4.1 SSE4.2 AVX2 FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62415/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62415/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62414	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62414/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62414/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62414/events	https://github.com/tensorflow/tensorflow/issues/62414	1,999,013,666	I_kwDOArmXAs53Joci	62,414	tf.keras.layers.AveragePooling3D crash	{ "login": "cheyennee", "id": 45327670, "node_id": "MDQ6VXNlcjQ1MzI3Njcw", "avatar_url": "https://avatars.githubusercontent.com/u/45327670?v=4", "gravatar_id": "", "url": "https://api.github.com/users/cheyennee", "html_url": "https://github.com/cheyennee", "followers_url": "https://api.github.com/users/cheyennee/followers", "following_url": "https://api.github.com/users/cheyennee/following{/other_user}", "gists_url": "https://api.github.com/users/cheyennee/gists{/gist_id}", "starred_url": "https://api.github.com/users/cheyennee/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/cheyennee/subscriptions", "organizations_url": "https://api.github.com/users/cheyennee/orgs", "repos_url": "https://api.github.com/users/cheyennee/repos", "events_url": "https://api.github.com/users/cheyennee/events{/privacy}", "received_events_url": "https://api.github.com/users/cheyennee/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097546578, "node_id": "MDU6TGFiZWwxMDk3NTQ2NTc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:keras", "name": "comp:keras", "color": "0052cc", "default": false, "description": "Keras related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false }	[ { "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false } ]	null	[ "@cheyennee,\r\n\r\nLooks like this is a duplicate of the issue [#56833](https://github.com/tensorflow/tensorflow/issues/56833). Could you please check and let us know if it is the same? Thank you!\r\n", "Hi, @tilakrayal,\r\nI believe these issues involve distinct factors. Specifically, in this issues, the error arises due to setting the strides to 0 in tf.keras.layers.AveragePooling3D. It's my understanding that the strides parameter in this context must be greater than 0; a stride of 0 is not permitted. Conversely, in issue #56833, the parameters used in tf.keras.layers.MaxPooling3D are all valid." ]	2023-11-17T12:27:55	2023-11-28T10:18:26	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14.0 ### Custom code Yes ### OS platform and distribution windows colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I run this code in colab, the colab crash and throw error message: ``` NotFoundError Traceback (most recent call last) [<ipython-input-5-afbc98afe688>](https://localhost:8080/#) in <cell line: 15>() 13 __input___0 = tf.identity(__input___0_tensor) 14 AveragePooling3D_class = tf.keras.layers.AveragePooling3D(pool_size=pool_size, strides=strides, padding=padding, data_format=data_format, dtype=tf.float64) ---> 15 print(AveragePooling3D_class(__input___0)) 16 # def wrap_func_exec(): 17 # pool_size_0 = 3 1 frames [/usr/local/lib/python3.10/dist-packages/tensorflow/python/eager/execute.py](https://localhost:8080/#) in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name) 58 for t in inputs 59 ] ---> 60 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name, 61 inputs, attrs, num_outputs) 62 except core._NotOkStatusException as e: NotFoundError: Exception encountered when calling layer 'average_pooling3d_6' (type AveragePooling3D). Could not find device for node: {{node AvgPool3D}} = AvgPool3D[T=DT_DOUBLE, data_format="NDHWC", ksize=[1, 3, 3, 3, 1], padding="VALID", strides=[1, 0, 21, 1, 1]] All kernels registered for op AvgPool3D: device='XLA_GPU_JIT'; T in [DT_FLOAT, DT_DOUBLE, DT_BFLOAT16, DT_HALF] device='XLA_CPU_JIT'; T in [DT_FLOAT, DT_DOUBLE, DT_BFLOAT16, DT_HALF] device='CPU'; T in [DT_BFLOAT16] device='CPU'; T in [DT_FLOAT] device='GPU'; T in [DT_BFLOAT16] device='GPU'; T in [DT_HALF] device='GPU'; T in [DT_FLOAT] [Op:AvgPool3D] Call arguments received by layer 'average_pooling3d_6' (type AveragePooling3D): • inputs=tf.Tensor(shape=(1, 11, 1, 1, 1), dtype=float64) ``` I also run this code in pycharm, the pycharm exits directly and ```the pycharm process finished with exit code -1073741819 (0xC0000005)``` ### Standalone code to reproduce the issue ```shell import tensorflow as tf pool_size_0 = 3 pool_size_1 = 3 pool_size_2 = 3 pool_size = [pool_size_0,pool_size_1,pool_size_2,] strides_0 = 0 strides_1 = 21.0 strides_2 = 1 strides = [strides_0,strides_1,strides_2,] padding = "valid" data_format = "channels_last" __input___0_tensor = tf.random.uniform([1, 11, 1, 1, 1], minval=-1.0, maxval=0.7764022115238914, dtype=tf.float32) __input___0 = tf.identity(__input___0_tensor) AveragePooling3D_class = tf.keras.layers.AveragePooling3D(pool_size=pool_size, strides=strides, padding=padding, data_format=data_format, dtype=tf.float64) print(AveragePooling3D_class(__input___0)) ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62414/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62414/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62413	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62413/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62413/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62413/events	https://github.com/tensorflow/tensorflow/issues/62413	1,998,369,859	I_kwDOArmXAs53HLRD	62,413	Will Tflite support flatbuffers v23.5.26 in the future ?(in this version, flatbuffers already support 64 bit field)	{ "login": "zhangguog", "id": 91189850, "node_id": "MDQ6VXNlcjkxMTg5ODUw", "avatar_url": "https://avatars.githubusercontent.com/u/91189850?v=4", "gravatar_id": "", "url": "https://api.github.com/users/zhangguog", "html_url": "https://github.com/zhangguog", "followers_url": "https://api.github.com/users/zhangguog/followers", "following_url": "https://api.github.com/users/zhangguog/following{/other_user}", "gists_url": "https://api.github.com/users/zhangguog/gists{/gist_id}", "starred_url": "https://api.github.com/users/zhangguog/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/zhangguog/subscriptions", "organizations_url": "https://api.github.com/users/zhangguog/orgs", "repos_url": "https://api.github.com/users/zhangguog/repos", "events_url": "https://api.github.com/users/zhangguog/events{/privacy}", "received_events_url": "https://api.github.com/users/zhangguog/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 1661751498, "node_id": "MDU6TGFiZWwxNjYxNzUxNDk4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TFLiteConverter", "name": "TFLiteConverter", "color": "bfdadc", "default": false, "description": "For issues related to TFLite converter" }, { "id": 2788658540, "node_id": "MDU6TGFiZWwyNzg4NjU4NTQw", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite-support", "name": "comp:lite-support", "color": "4A5F20", "default": false, "description": "Lite support library related issues" } ]	closed	false	{ "login": "zichuan-wei", "id": 10928342, "node_id": "MDQ6VXNlcjEwOTI4MzQy", "avatar_url": "https://avatars.githubusercontent.com/u/10928342?v=4", "gravatar_id": "", "url": "https://api.github.com/users/zichuan-wei", "html_url": "https://github.com/zichuan-wei", "followers_url": "https://api.github.com/users/zichuan-wei/followers", "following_url": "https://api.github.com/users/zichuan-wei/following{/other_user}", "gists_url": "https://api.github.com/users/zichuan-wei/gists{/gist_id}", "starred_url": "https://api.github.com/users/zichuan-wei/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/zichuan-wei/subscriptions", "organizations_url": "https://api.github.com/users/zichuan-wei/orgs", "repos_url": "https://api.github.com/users/zichuan-wei/repos", "events_url": "https://api.github.com/users/zichuan-wei/events{/privacy}", "received_events_url": "https://api.github.com/users/zichuan-wei/received_events", "type": "User", "site_admin": false }	[ { "login": "zichuan-wei", "id": 10928342, "node_id": "MDQ6VXNlcjEwOTI4MzQy", "avatar_url": "https://avatars.githubusercontent.com/u/10928342?v=4", "gravatar_id": "", "url": "https://api.github.com/users/zichuan-wei", "html_url": "https://github.com/zichuan-wei", "followers_url": "https://api.github.com/users/zichuan-wei/followers", "following_url": "https://api.github.com/users/zichuan-wei/following{/other_user}", "gists_url": "https://api.github.com/users/zichuan-wei/gists{/gist_id}", "starred_url": "https://api.github.com/users/zichuan-wei/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/zichuan-wei/subscriptions", "organizations_url": "https://api.github.com/users/zichuan-wei/orgs", "repos_url": "https://api.github.com/users/zichuan-wei/repos", "events_url": "https://api.github.com/users/zichuan-wei/events{/privacy}", "received_events_url": "https://api.github.com/users/zichuan-wei/received_events", "type": "User", "site_admin": false }, { "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @zichuan-wei, do you know the answer to this? Thanks for your help.", "Hi @zhangguog , \r\nThanks for the question, so far we don't have plans for migrating over to Flatbuffer 64 bit fields. We do however continue to evaluate different solutions and will revisit this if a need for 64bit fields arises" ]	2023-11-17T06:30:51	2023-11-21T22:51:38	2023-11-21T22:51:11	NONE	null	null	null	### 1. System information Linux ### 2. Question I already see that Tesnorflow Lite can support models > 2GB in schema Version 3c, but I also see that Flatbuffers add "offset64" and "vector64" in latest version. So I want to ask if Tesnorflow Lite will use Flatbuffers 64 bit field instead of schema Version 3c' upgrade?	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62413/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62413/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62412	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62412/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62412/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62412/events	https://github.com/tensorflow/tensorflow/issues/62412	1,997,303,844	I_kwDOArmXAs53DHAk	62,412	Tensorflow 2.15 Docker image cannot find the GPU drivers, but nvidia-smi can.	{ "login": "gmarcusm", "id": 4169784, "node_id": "MDQ6VXNlcjQxNjk3ODQ=", "avatar_url": "https://avatars.githubusercontent.com/u/4169784?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gmarcusm", "html_url": "https://github.com/gmarcusm", "followers_url": "https://api.github.com/users/gmarcusm/followers", "following_url": "https://api.github.com/users/gmarcusm/following{/other_user}", "gists_url": "https://api.github.com/users/gmarcusm/gists{/gist_id}", "starred_url": "https://api.github.com/users/gmarcusm/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gmarcusm/subscriptions", "organizations_url": "https://api.github.com/users/gmarcusm/orgs", "repos_url": "https://api.github.com/users/gmarcusm/repos", "events_url": "https://api.github.com/users/gmarcusm/events{/privacy}", "received_events_url": "https://api.github.com/users/gmarcusm/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547538, "node_id": "MDU6TGFiZWwxMDk3NTQ3NTM4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:gpu", "name": "comp:gpu", "color": "0052cc", "default": false, "description": "GPU related issues" }, { "id": 6218999181, "node_id": "LA_kwDOArmXAs8AAAABcq5ljQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.15", "name": "TF 2.15", "color": "9162CB", "default": false, "description": "For issues related to 2.15.x" } ]	open	false	{ "login": "MichaelHudgins", "id": 30155094, "node_id": "MDQ6VXNlcjMwMTU1MDk0", "avatar_url": "https://avatars.githubusercontent.com/u/30155094?v=4", "gravatar_id": "", "url": "https://api.github.com/users/MichaelHudgins", "html_url": "https://github.com/MichaelHudgins", "followers_url": "https://api.github.com/users/MichaelHudgins/followers", "following_url": "https://api.github.com/users/MichaelHudgins/following{/other_user}", "gists_url": "https://api.github.com/users/MichaelHudgins/gists{/gist_id}", "starred_url": "https://api.github.com/users/MichaelHudgins/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/MichaelHudgins/subscriptions", "organizations_url": "https://api.github.com/users/MichaelHudgins/orgs", "repos_url": "https://api.github.com/users/MichaelHudgins/repos", "events_url": "https://api.github.com/users/MichaelHudgins/events{/privacy}", "received_events_url": "https://api.github.com/users/MichaelHudgins/received_events", "type": "User", "site_admin": false }	[ { "login": "MichaelHudgins", "id": 30155094, "node_id": "MDQ6VXNlcjMwMTU1MDk0", "avatar_url": "https://avatars.githubusercontent.com/u/30155094?v=4", "gravatar_id": "", "url": "https://api.github.com/users/MichaelHudgins", "html_url": "https://github.com/MichaelHudgins", "followers_url": "https://api.github.com/users/MichaelHudgins/followers", "following_url": "https://api.github.com/users/MichaelHudgins/following{/other_user}", "gists_url": "https://api.github.com/users/MichaelHudgins/gists{/gist_id}", "starred_url": "https://api.github.com/users/MichaelHudgins/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/MichaelHudgins/subscriptions", "organizations_url": "https://api.github.com/users/MichaelHudgins/orgs", "repos_url": "https://api.github.com/users/MichaelHudgins/repos", "events_url": "https://api.github.com/users/MichaelHudgins/events{/privacy}", "received_events_url": "https://api.github.com/users/MichaelHudgins/received_events", "type": "User", "site_admin": false }, { "login": "angerson", "id": 32465472, "node_id": "MDQ6VXNlcjMyNDY1NDcy", "avatar_url": "https://avatars.githubusercontent.com/u/32465472?v=4", "gravatar_id": "", "url": "https://api.github.com/users/angerson", "html_url": "https://github.com/angerson", "followers_url": "https://api.github.com/users/angerson/followers", "following_url": "https://api.github.com/users/angerson/following{/other_user}", "gists_url": "https://api.github.com/users/angerson/gists{/gist_id}", "starred_url": "https://api.github.com/users/angerson/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/angerson/subscriptions", "organizations_url": "https://api.github.com/users/angerson/orgs", "repos_url": "https://api.github.com/users/angerson/repos", "events_url": "https://api.github.com/users/angerson/events{/privacy}", "received_events_url": "https://api.github.com/users/angerson/received_events", "type": "User", "site_admin": false }, { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "Hello @gmarcusm!\r\nIf you are not using TF v2.15 specifically then, could you try to use TF v2.14 which willl work fine with TF GPU drivers. Thank you!", "> Hello @gmarcusm! If you are not using TF v2.15 specifically then, could you try to use TF v2.14 which willl work fine with TF GPU drivers. Thank you!\r\n\r\nHi ! I use this as part of a CI workflow and I want to test against TF 2.15 specifically, so it does require 2.15 GPU support in this case.\r\n\r\nIf the `-gpu` docker images are not working as intended, shouldn't they be removed until the issue is fixed?", "I am using TF 2.15 with cuda 12.3, and I have encountered the same problem that my tf cannot find GPUs while nvidia-smi can.", "I have the same issue after updated v1.14.3 nvidia-container-toolkit. \r\nGPU 2.14 and nightly work fine.\r\n```console\r\nwilliamharris@pop-os:~$ docker run --gpus all -it tensorflow/tensorflow:2.15.0-gpu-jupyter bash\r\n\r\n________ _______________\r\n___ __/__________________________________ ____/__ /________ __\r\n__ / _ _ \\_ __ \\_ ___/ __ \\_ ___/_ /_ __ /_ __ \\_ \| /\| / /\r\n_ / / __/ / / /(__ )/ /_/ / / _ __/ _ / / /_/ /_ \|/ \|/ /\r\n/_/ \\___//_/ /_//____/ \\____//_/ /_/ /_/ \\____/____/\|__/\r\n\r\n\r\nWARNING: You are running this container as root, which can cause new files in\r\nmounted volumes to be created as the root user on your host machine.\r\n\r\nTo avoid this, run the container by specifying your user's userid:\r\n\r\n$ docker run -u $(id -u):$(id -g) args...\r\n\r\nroot@087d48984615:/tf# nvidia-smi\r\nTue Nov 21 05:53:18 2023 \r\n+---------------------------------------------------------------------------------------+\r\n\| NVIDIA-SMI 545.29.02 Driver Version: 545.29.02 CUDA Version: 12.3 \|\r\n\|-----------------------------------------+----------------------+----------------------+\r\n\| GPU Name Persistence-M \| Bus-Id Disp.A \| Volatile Uncorr. ECC \|\r\n\| Fan Temp Perf Pwr:Usage/Cap \| Memory-Usage \| GPU-Util Compute M. \|\r\n\| \| \| MIG M. \|\r\n\|=========================================+======================+======================\|\r\n\| 0 NVIDIA GeForce RTX 2070 ... Off \| 00000000:01:00.0 On \| N/A \|\r\n\| 0% 41C P8 26W / 235W \| 507MiB / 8192MiB \| 3% Default \|\r\n\| \| \| N/A \|\r\n+-----------------------------------------+----------------------+----------------------+\r\n \r\n+---------------------------------------------------------------------------------------+\r\n\| Processes: \|\r\n\| GPU GI CI PID Type Process name GPU Memory \|\r\n\| ID ID Usage \|\r\n\|=======================================================================================\|\r\n+---------------------------------------------------------------------------------------+\r\nroot@087d48984615:/tf# python3 -c \"import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))\"\r\n2023-11-21 05:53:26.043780: I tensorflow/core/util/port.cc:113] oneDNN custom operations are on. You may see slightly different numerical results due to floating-point round-off errors from different computation orders. To turn them off, set the environment variable `TF_ENABLE_ONEDNN_OPTS=0`.\r\n2023-11-21 05:53:26.044955: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.\r\n2023-11-21 05:53:26.062586: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered\r\n2023-11-21 05:53:26.062604: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered\r\n2023-11-21 05:53:26.063167: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered\r\n2023-11-21 05:53:26.066161: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used.\r\n2023-11-21 05:53:26.066261: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations.\r\nTo enable the following instructions: AVX2 AVX_VNNI FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags.\r\n2023-11-21 05:53:26.636830: I external/local_xla/xla/stream_executor/cuda/cuda_executor.cc:901] successful NUMA node read from SysFS had negative value (-1), but there must be at least one NUMA node, so returning NUMA node zero. See more at https://github.com/torvalds/linux/blob/v6.0/Documentation/ABI/testing/sysfs-bus-pci#L344-L355\r\n2023-11-21 05:53:26.637345: W tensorflow/core/common_runtime/gpu/gpu_device.cc:2256] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform.\r\nSkipping registering GPU devices...\r\n[]\r\nroot@087d48984615:/tf# \r\n```", "I think I have the same problem.\r\nGPU: A800 DriveVersion: 525.105.17 CUDA Version: 12.0\r\nI’ve installed such versions below\r\ntensorflow 2.9.0 / 2.12/ 2.13.1 / 2.14.0\r\n\r\nBut it seems like tensorflow could not detect GPU because:\r\navailable = tf.test.is_gpu_available() is_cuda_gpu_available = tf.test.is_gpu_available(cuda_only=True) is_cuda_gpu_min_3 = tf.test.is_gpu_available(True, (3,0)) All were “False”.\r\n\r\nWhen I tried pip install tensorflow[and-cuda]==2.15.0 It exactly returns “True”\r\n\r\nBut still received the error message:\r\n`E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered`", "@angerson could you PTAL when able?", "For using pip to install in Colab, please follow the below step for GPU to be detected in Tensorflow 2.15, till the issue is resolved.\r\n\r\n`pip install --extra-index-url https://pypi.nvidia.com tensorflow[and-cuda]` works for upgrading to TF 2.15 and detects GPU in Colab.", "Hello,\r\n\r\nI came across this thread and noticed that I am experiencing a similar issue with TensorFlow 2.15 on my system. I have been unable to get TensorFlow to recognize my GPU, and I thought sharing my setup and steps I've taken might contribute to finding a solution.\r\n\r\nHere are the specifics of my setup:\r\n\r\nOperating System: Windows 11 Home\r\nPython Version: 3.10.1\r\nTensorFlow Version: 2.15\r\nCUDA Version: 12.3\r\ncuDNN Version: 12\r\nGPU: NVIDIA GeForce RTX 4070 Ti\r\nGPU Driver Version: (Please include your GPU driver version here)\r\nDespite correctly setting up the environment variables and restarting my computer after each configuration, tf.config.list_physical_devices('GPU') returns an empty list, and testing cuDNN installation returns false.\r\n\r\nIt's interesting to note that this problem persists with TensorFlow 2.15, as I have taken all recommended steps to ensure proper installation and configuration. If anyone has insights or suggestions on what might be going wrong, it would be greatly appreciated.\r\n\r\nThank you for your time and help!\r\n\r\nBest regards,\r\nbayerno(helped by GPT 4)", "I encountered the same problem.\r\n```python\r\ntf.__version__ # '2.15.0'\r\nprint(tf.config.list_physical_devices('GPU')) # []\r\n```\r\n\r\nnvidia-smi\r\n```bash\r\nSat Feb 3 18:53:33 2024 \r\n+---------------------------------------------------------------------------------------+\r\n\| NVIDIA-SMI 546.12 Driver Version: 546.12 CUDA Version: 12.3 \|\r\n\|-----------------------------------------+----------------------+----------------------+\r\n\| GPU Name TCC/WDDM \| Bus-Id Disp.A \| Volatile Uncorr. ECC \|\r\n\| Fan Temp Perf Pwr:Usage/Cap \| Memory-Usage \| GPU-Util Compute M. \|\r\n\| \| \| MIG M. \|\r\n\|=========================================+======================+======================\|\r\n\| 0 NVIDIA GeForce RTX 4060 ... WDDM \| 00000000:01:00.0 Off \| N/A \|\r\n\| N/A 44C P0 14W / 115W \| 0MiB / 8188MiB \| 0% Default \|\r\n\| \| \| N/A \|\r\n+-----------------------------------------+----------------------+----------------------+\r\n\r\n+---------------------------------------------------------------------------------------+\r\n\| Processes: \|\r\n\| GPU GI CI PID Type Process name GPU Memory \|\r\n\| ID ID Usage \|\r\n\|=======================================================================================\|\r\n\| No running processes found \|\r\n+---------------------------------------------------------------------------------------+\r\n```\r\n" ]	2023-11-16T16:52:27	2024-04-20T21:25:46	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source binary ### TensorFlow version TF 2.15.0 ### Custom code No ### OS platform and distribution Linux Ubuntu 22.04 ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version CUDA 12.2 ### GPU model and memory NVIDIA TITAN V ### Current behavior? Running TensorFlow 2.15.0 from within Docker image does not find GPU drivers. nvidia-smi reports the GPUs as available. This works as intended with a TF 2.14 image in the same machine. ### Standalone code to reproduce the issue ```shell docker run --gpus all -it tensorflow/tensorflow:2.15.0-gpu-jupyter bash <within the container> # nvidia-smi # python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))" ``` ### Relevant log output ```shell # nvidia-smi Thu Nov 16 16:47:22 2023 +---------------------------------------------------------------------------------------+ \| NVIDIA-SMI 535.129.03 Driver Version: 535.129.03 CUDA Version: 12.2 \| \|-----------------------------------------+----------------------+----------------------+ \| GPU Name Persistence-M \| Bus-Id Disp.A \| Volatile Uncorr. ECC \| \| Fan Temp Perf Pwr:Usage/Cap \| Memory-Usage \| GPU-Util Compute M. \| \| \| \| MIG M. \| \|=========================================+======================+======================\| \| 0 NVIDIA TITAN V Off \| 00000000:65:00.0 Off \| N/A \| \| 29% 46C P8 27W / 250W \| 1693MiB / 12288MiB \| 2% Default \| \| \| \| N/A \| +-----------------------------------------+----------------------+----------------------+ +---------------------------------------------------------------------------------------+ \| Processes: \| \| GPU GI CI PID Type Process name GPU Memory \| \| ID ID Usage \| \|=======================================================================================\| +---------------------------------------------------------------------------------------+ # python3 -c "import tensorflow as tf; print(tf.config.list_physical_devices('GPU'))" 2023-11-16 16:46:54.131081: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used. 2023-11-16 16:46:54.255566: E external/local_xla/xla/stream_executor/cuda/cuda_dnn.cc:9261] Unable to register cuDNN factory: Attempting to register factory for plugin cuDNN when one has already been registered 2023-11-16 16:46:54.255623: E external/local_xla/xla/stream_executor/cuda/cuda_fft.cc:607] Unable to register cuFFT factory: Attempting to register factory for plugin cuFFT when one has already been registered 2023-11-16 16:46:54.276648: E external/local_xla/xla/stream_executor/cuda/cuda_blas.cc:1515] Unable to register cuBLAS factory: Attempting to register factory for plugin cuBLAS when one has already been registered 2023-11-16 16:46:54.327586: I external/local_tsl/tsl/cuda/cudart_stub.cc:31] Could not find cuda drivers on your machine, GPU will not be used. 2023-11-16 16:46:54.328299: I tensorflow/core/platform/cpu_feature_guard.cc:182] This TensorFlow binary is optimized to use available CPU instructions in performance-critical operations. To enable the following instructions: AVX2 AVX512F FMA, in other operations, rebuild TensorFlow with the appropriate compiler flags. 2023-11-16 16:46:56.486851: W tensorflow/core/common_runtime/gpu/gpu_device.cc:2256] Cannot dlopen some GPU libraries. Please make sure the missing libraries mentioned above are installed properly if you would like to use GPU. Follow the guide at https://www.tensorflow.org/install/gpu for how to download and setup the required libraries for your platform. Skipping registering GPU devices... [] ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62412/reactions", "total_count": 13, "+1": 13, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62412/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62411	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62411/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62411/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62411/events	https://github.com/tensorflow/tensorflow/pull/62411	1,997,108,510	PR_kwDOArmXAs5fpK_X	62,411	lite: add tensors and nodes size in SignatureRunner	{ "login": "aflaischer", "id": 26463144, "node_id": "MDQ6VXNlcjI2NDYzMTQ0", "avatar_url": "https://avatars.githubusercontent.com/u/26463144?v=4", "gravatar_id": "", "url": "https://api.github.com/users/aflaischer", "html_url": "https://github.com/aflaischer", "followers_url": "https://api.github.com/users/aflaischer/followers", "following_url": "https://api.github.com/users/aflaischer/following{/other_user}", "gists_url": "https://api.github.com/users/aflaischer/gists{/gist_id}", "starred_url": "https://api.github.com/users/aflaischer/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/aflaischer/subscriptions", "organizations_url": "https://api.github.com/users/aflaischer/orgs", "repos_url": "https://api.github.com/users/aflaischer/repos", "events_url": "https://api.github.com/users/aflaischer/events{/privacy}", "received_events_url": "https://api.github.com/users/aflaischer/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 1169364458, "node_id": "MDU6TGFiZWwxMTY5MzY0NDU4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:S", "name": "size:S", "color": "adafea", "default": false, "description": "CL Change Size: Small" } ]	open	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @majiddadashi Can you please review this PR ? Thank you!\r\n", "Hi @majiddadashi Can you please review this PR ? Thank you!", "Hi @majiddadashi Can you please review this PR ? Thank you!", "Hi @majiddadashi Can you please review this PR ? Thank you!", "hi @gbaned, is there something I can do regarding the pull request?", "> hi @gbaned, is there something I can do regarding the pull request?\r\n\r\nHi @aflaischer Sorry for the delay, this PR is awaiting review status and will be processed further once it is approved. Nothing is pending on your end. Thank you!", "Hi @majiddadashi Can you please review this PR ? Thank you!", "Hi @majiddadashi Can you please review this PR ? Thank you!" ]	2023-11-16T15:19:22	2024-06-07T16:35:21	null	NONE	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62411", "html_url": "https://github.com/tensorflow/tensorflow/pull/62411", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62411.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62411.patch", "merged_at": null }	number of tensors and nodes is already available in Subgraph class, map these sizes in SignatureRunner. It avoids to use GetSubgraphIndexFromSignature() and subgraph() from Interpreter class to retrieve these sizes for a given signature.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62411/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62411/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62410	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62410/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62410/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62410/events	https://github.com/tensorflow/tensorflow/issues/62410	1,996,932,295	I_kwDOArmXAs53BsTH	62,410	Using TensorFlow Lite for android development, copied repo from source and while running code I am facing error	{ "login": "kartik-waghmare", "id": 143528663, "node_id": "U_kgDOCI4S1w", "avatar_url": "https://avatars.githubusercontent.com/u/143528663?v=4", "gravatar_id": "", "url": "https://api.github.com/users/kartik-waghmare", "html_url": "https://github.com/kartik-waghmare", "followers_url": "https://api.github.com/users/kartik-waghmare/followers", "following_url": "https://api.github.com/users/kartik-waghmare/following{/other_user}", "gists_url": "https://api.github.com/users/kartik-waghmare/gists{/gist_id}", "starred_url": "https://api.github.com/users/kartik-waghmare/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/kartik-waghmare/subscriptions", "organizations_url": "https://api.github.com/users/kartik-waghmare/orgs", "repos_url": "https://api.github.com/users/kartik-waghmare/repos", "events_url": "https://api.github.com/users/kartik-waghmare/events{/privacy}", "received_events_url": "https://api.github.com/users/kartik-waghmare/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 1661751498, "node_id": "MDU6TGFiZWwxNjYxNzUxNDk4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TFLiteConverter", "name": "TFLiteConverter", "color": "bfdadc", "default": false, "description": "For issues related to TFLite converter" }, { "id": 4989164230, "node_id": "LA_kwDOArmXAs8AAAABKWCaxg", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/Android", "name": "Android", "color": "e99695", "default": false, "description": "" } ]	closed	false	{ "login": "LakshmiKalaKadali", "id": 149650845, "node_id": "U_kgDOCOt9nQ", "avatar_url": "https://avatars.githubusercontent.com/u/149650845?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LakshmiKalaKadali", "html_url": "https://github.com/LakshmiKalaKadali", "followers_url": "https://api.github.com/users/LakshmiKalaKadali/followers", "following_url": "https://api.github.com/users/LakshmiKalaKadali/following{/other_user}", "gists_url": "https://api.github.com/users/LakshmiKalaKadali/gists{/gist_id}", "starred_url": "https://api.github.com/users/LakshmiKalaKadali/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LakshmiKalaKadali/subscriptions", "organizations_url": "https://api.github.com/users/LakshmiKalaKadali/orgs", "repos_url": "https://api.github.com/users/LakshmiKalaKadali/repos", "events_url": "https://api.github.com/users/LakshmiKalaKadali/events{/privacy}", "received_events_url": "https://api.github.com/users/LakshmiKalaKadali/received_events", "type": "User", "site_admin": false }	[ { "login": "LakshmiKalaKadali", "id": 149650845, "node_id": "U_kgDOCOt9nQ", "avatar_url": "https://avatars.githubusercontent.com/u/149650845?v=4", "gravatar_id": "", "url": "https://api.github.com/users/LakshmiKalaKadali", "html_url": "https://github.com/LakshmiKalaKadali", "followers_url": "https://api.github.com/users/LakshmiKalaKadali/followers", "following_url": "https://api.github.com/users/LakshmiKalaKadali/following{/other_user}", "gists_url": "https://api.github.com/users/LakshmiKalaKadali/gists{/gist_id}", "starred_url": "https://api.github.com/users/LakshmiKalaKadali/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/LakshmiKalaKadali/subscriptions", "organizations_url": "https://api.github.com/users/LakshmiKalaKadali/orgs", "repos_url": "https://api.github.com/users/LakshmiKalaKadali/repos", "events_url": "https://api.github.com/users/LakshmiKalaKadali/events{/privacy}", "received_events_url": "https://api.github.com/users/LakshmiKalaKadali/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @kartik-waghmare, \r\n\r\nThe Issuelog displaying here might be due to gradle version. \r\nPlease provide some more context to better understand the issue and Which Tflite example you have considered? \r\n\r\n\r\nThank You", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "I've the same issues and here is my Gradle version.\r\n\r\n~/StudioProjects/examples]: \r\n → ./lite/examples/object_detection/android_play_services/gradlew --version\r\n\r\nWelcome to Gradle 8.4!\r\n\r\nHere are the highlights of this release:\r\n - Compiling and testing with Java 21\r\n - Faster Java compilation on Windows\r\n - Role focused dependency configurations creation\r\n\r\nFor more details see https://docs.gradle.org/8.4/release-notes.html\r\n\r\n\r\n------------------------------------------------------------\r\nGradle 8.4\r\n------------------------------------------------------------\r\n\r\nBuild time: 2023-10-04 20:52:13 UTC\r\nRevision: e9251e572c9bd1d01e503a0dfdf43aedaeecdc3f\r\n\r\nKotlin: 1.9.10\r\nGroovy: 3.0.17\r\nAnt: Apache Ant(TM) version 1.10.13 compiled on January 4 2023\r\nJVM: 11.0.21 (Debian 11.0.21+9-post-Debian-1)\r\nOS: Linux 6.5.6-1rodete4-amd64 amd64\r\n", "same issue for me\r\ndistributionBase=GRADLE_USER_HOME\r\ndistributionPath=wrapper/dists\r\ndistributionUrl=https\\://services.gradle.org/distributions/gradle-8.4-bin.zip\r\nnetworkTimeout=10000\r\nvalidateDistributionUrl=true\r\nzipStoreBase=GRADLE_USER_HOME\r\nzipStorePath=wrapper/dists\r\n\r\nI changed gradle version and it worked\r\ndistributionBase=GRADLE_USER_HOME\r\ndistributionPath=wrapper/dists\r\nzipStoreBase=GRADLE_USER_HOME\r\nzipStorePath=wrapper/dists\r\ndistributionUrl=https\\://services.gradle.org/distributions/gradle-7.3.3-bin.zip" ]	2023-11-16T14:04:05	2024-03-22T10:34:12	2023-12-09T01:48:16	NONE	null	null	null	### 1. System information - OS Platform and Distribution (e.g., Linux Ubuntu 16.04): Microsoft Windows 10 Pro - TensorFlow library (version, if pip package or github SHA, if built from source): git clone https://github.com/tensorflow/examples.git ### 2. Software Android Studio Flamingo \| 2022.2.1 Patch 2 Issue Log: Cannot use @TaskAction annotation on method DataBindingGenBaseClassesTask.writeBaseClasses() because interface org.gradle.api.tasks.incremental.IncrementalTaskInputs is not a valid parameter to an action method	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62410/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62410/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62409	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62409/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62409/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62409/events	https://github.com/tensorflow/tensorflow/pull/62409	1,996,223,649	PR_kwDOArmXAs5fmIXp	62,409	Register complex dtypes for TruncateDiv Op	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "id": 1169364259, "node_id": "MDU6TGFiZWwxMTY5MzY0MjU5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:XS", "name": "size:XS", "color": "adafea", "default": false, "description": "CL Change Size: Extra Small" }, { "id": 1478826728, "node_id": "MDU6TGFiZWwxNDc4ODI2NzI4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:core", "name": "comp:core", "color": "024391", "default": false, "description": "issues related to core part of tensorflow" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "> As per documentation the Op tf.truncatediv should support complex dtypes also.\r\n\r\nThank you for noticing this. Floor/truncate division isn't supported for complex types, so the documentation is wrong.\r\n", "Hi @SuryanarayanaY As mentioned by reviewers in above comments, Floor/truncate division isn't supported for complex types. Hence closing this PR. Thank you!" ]	2023-11-16T07:21:53	2024-03-07T16:48:29	2024-03-07T08:15:57	COLLABORATOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62409", "html_url": "https://github.com/tensorflow/tensorflow/pull/62409", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62409.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62409.patch", "merged_at": null }	As per documentation the Op `tf.truncatediv` should support `complex` dtypes also. > Args > -- > x \| A Tensor. Must be one of the following types: bfloat16, half, float32, float64, uint8, int8, uint16, int16, int32, uint32, uint64, int64, complex64, complex128. As per the `math_ops.cc` , `complex` dtypes also registered for this Op. https://github.com/tensorflow/tensorflow/blob/e63d37e07b95e5c6e8a78794547fe317454edae7/tensorflow/core/ops/math_ops.cc#L521-L523 https://github.com/tensorflow/tensorflow/blob/e63d37e07b95e5c6e8a78794547fe317454edae7/tensorflow/core/ops/math_ops.cc#L392-L395 Hence adding the complex dtypes also to this Op registry.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62409/reactions", "total_count": 1, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 1, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62409/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62408	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62408/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62408/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62408/events	https://github.com/tensorflow/tensorflow/pull/62408	1,996,175,001	PR_kwDOArmXAs5fl9w7	62,408	Update supported dtypes of truncatediv Op in gpu_supported_ops.md	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169364259, "node_id": "MDU6TGFiZWwxMTY5MzY0MjU5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:XS", "name": "size:XS", "color": "adafea", "default": false, "description": "CL Change Size: Extra Small" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @cheshire Can you please review this PR ? Thank you!" ]	2023-11-16T06:44:42	2023-12-16T03:30:54	2023-12-16T03:30:54	COLLABORATOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62408", "html_url": "https://github.com/tensorflow/tensorflow/pull/62408", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62408.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62408.patch", "merged_at": "2023-12-16T03:30:53" }	Updated supported `dtypes` of `truncatediv` Op which is missing the list. From: `{complex64,double,float,int32,int64} ` To: `{complex64,complex128,double,float,half,bfloat16,int8,int16,int32,int64,uint8,uint16,uint32,uint64} ` as this Op also supports mentioned dtypes. May please refer attached [gist](https://colab.sandbox.google.com/gist/SuryanarayanaY/b0e584a4d45599bbff914d4119640d8e/tf-truncatediv_with_jit.ipynb).	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62408/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62408/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62407	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62407/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62407/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62407/events	https://github.com/tensorflow/tensorflow/issues/62407	1,995,861,895	I_kwDOArmXAs529m-H	62,407	TensorFlow Lite in Play Services issue: Cannot initialize interpreter: Error loading TFLite native library	{ "login": "jahonn", "id": 1058051, "node_id": "MDQ6VXNlcjEwNTgwNTE=", "avatar_url": "https://avatars.githubusercontent.com/u/1058051?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jahonn", "html_url": "https://github.com/jahonn", "followers_url": "https://api.github.com/users/jahonn/followers", "following_url": "https://api.github.com/users/jahonn/following{/other_user}", "gists_url": "https://api.github.com/users/jahonn/gists{/gist_id}", "starred_url": "https://api.github.com/users/jahonn/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/jahonn/subscriptions", "organizations_url": "https://api.github.com/users/jahonn/orgs", "repos_url": "https://api.github.com/users/jahonn/repos", "events_url": "https://api.github.com/users/jahonn/events{/privacy}", "received_events_url": "https://api.github.com/users/jahonn/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473184161, "node_id": "MDU6TGFiZWw0NzMxODQxNjE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:support", "name": "type:support", "color": "159b2e", "default": false, "description": "Support issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 2671339633, "node_id": "MDU6TGFiZWwyNjcxMzM5NjMz", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TFLiteGpuDelegate", "name": "TFLiteGpuDelegate", "color": "F71F04", "default": false, "description": "TFLite Gpu delegate issue" }, { "id": 3835861157, "node_id": "LA_kwDOArmXAs7kopil", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TFLiteGooglePlayServices", "name": "TFLiteGooglePlayServices", "color": "27FAE2", "default": false, "description": "For issues related to TensorFlow Lite in Google Play Services" } ]	closed	false	{ "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false }	[ { "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false } ]	null	[ "@jahonn,\r\nCould you please try to add the below line in this case\r\n\r\n` implementation 'com.google.android.gms:play-services-tflite-java:16.1.0 `\r\n\r\nto the Tensorflow lite dependencies section \r\n\r\n(OTOH, you might not need the `org.tensorflow:tensorflow-lite-task-vision-play-services` dependency if you are using InterpreterApi. Also please take a look at the [Interpreter API dependencies](https://www.tensorflow.org/lite/android/play_services#1_add_project_dependencies_2). Thank you!", "Hi @tilakrayal \r\n\r\nI have already added the Tensorflow Lite dependencies in my build.gradle, and got the error.\r\n\r\nIt seems the root error is\r\n`\r\n java.lang.UnsatisfiedLinkError: dlopen failed: couldn't map \"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\" segment 1: Permission denied\r\n`\r\n\r\nYes, I'm using InterpreterApi.", "I opened the dlerror and dlopen log, and get this.\r\n\r\n> \r\n\r\n11-17 10:14:03.807 17066 17688 D linker : dlopen(name=\"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\", flags=0x2, extinfo=[flags=0x200, reserved_addr=0x0, reserved_size=0x0, relro_fd=0, library_fd=0, library_fd_offset=0x0, library_namespace=classloader-namespace@0x76152ccf10], caller=\"/apex/com.android.art/lib64/libnativeloader.so\", caller_ns=com_android_art@0x76152cc1f0, targetSdkVersion=33) ...\r\n11-17 10:14:03.808 17066 17688 D linker : find_libraries(ns=classloader-namespace): task=/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so, is_dt_needed=0\r\n11-17 10:14:03.808 17066 17688 D linker : load_library(ns=classloader-namespace, task=/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so, flags=0x2, search_linked_namespaces=1): calling open_library\r\n11-17 10:14:03.808 17066 17688 D linker : load_library(ns=classloader-namespace, task=/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so, flags=0x2, realpath=/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so, search_linked_namespaces=1)\r\n11-17 10:14:03.809 17066 17688 D linker : load_library(ns=classloader-namespace, task=/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so): Adding DT_NEEDED task: libdl.so\r\n11-17 10:14:03.809 17066 17688 D linker : load_library(ns=classloader-namespace, task=/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so): Adding DT_NEEDED task: libm.so\r\n11-17 10:14:03.809 17066 17688 D linker : load_library(ns=classloader-namespace, task=/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so): Adding DT_NEEDED task: liblog.so\r\n11-17 10:14:03.809 17066 17688 D linker : load_library(ns=classloader-namespace, task=/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so): Adding DT_NEEDED task: libc.so\r\n11-17 10:14:03.809 17066 17688 D linker : find_libraries(ns=classloader-namespace): task=libdl.so, is_dt_needed=1\r\n11-17 10:14:03.809 17066 17688 D linker : find_library_internal(ns=classloader-namespace, task=libdl.so): Already loaded (by soname): /apex/com.android.runtime/lib64/bionic/libdl.so\r\n11-17 10:14:03.809 17066 17688 D linker : find_libraries(ns=classloader-namespace): task=libm.so, is_dt_needed=1\r\n11-17 10:14:03.809 17066 17688 D linker : find_library_internal(ns=classloader-namespace, task=libm.so): Already loaded (by soname): /apex/com.android.runtime/lib64/bionic/libm.so\r\n11-17 10:14:03.809 17066 17688 D linker : find_libraries(ns=classloader-namespace): task=liblog.so, is_dt_needed=1\r\n11-17 10:14:03.810 17066 17688 D linker : find_library_internal(ns=classloader-namespace, task=liblog.so): Already loaded (by soname): /system/lib64/liblog.so\r\n11-17 10:14:03.810 17066 17688 D linker : find_libraries(ns=classloader-namespace): task=libc.so, is_dt_needed=1\r\n11-17 10:14:03.810 17066 17688 D linker : find_library_internal(ns=classloader-namespace, task=libc.so): Already loaded (by soname): /apex/com.android.runtime/lib64/bionic/libc.so\r\n11-17 10:14:03.810 17066 17688 D linker : ... dlclose(realpath=\"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\"@0x76153c8598) ... load group root is \"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\"@0x76153c8598\r\n11-17 10:14:03.808 17066 17066 W TFLiteClient-0: type=1400 audit(0.0:11423): avc: denied { execute } for path=\"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk\" dev=\"dm-41\" ino=20657 scontext=u:r:platform_app:s0:c512,c768 tcontext=u:object_r:privapp_data_file:s0:c512,c768 tclass=file permissive=0 app=xxxxxxxxxxx\r\n11-17 10:14:03.810 17066 17688 D linker : ... dlclose: calling destructors for \"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\"@0x76153c8598 ...\r\n11-17 10:14:03.810 17066 17688 D linker : ... dlclose: calling destructors for \"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\"@0x76153c8598 ... done\r\n11-17 10:14:03.811 17066 17688 D linker : ... dlclose: unloading \"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\"@0x76153c8598 ...\r\n11-17 10:14:03.811 17066 17688 D linker : ... dlclose: unload_si was not linked - not unloading external references ...\r\n11-17 10:14:03.811 17066 17688 D linker : ... dlopen failed: couldn't map \"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\" segment 1: Permission denied\r\n11-17 10:14:03.811 17066 17688 D linker : dlerror set to \"dlopen failed: couldn't map \"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\" segment 1: Permission denied\"", "@pkgoogle, \r\nPlease look into the issue\r\n\r\nThank you", "Hi @jahonn, we need more context to reproduce this, If you can export your android project into a zip file that will be easiest. If you are following a resource please let us know which resource it is. If it contains proprietary code, see if you can start a new project and make the minimal changes/additions to reproduce the issue, then export this project. Thanks for your help.", "hi @pkgoogle ,\r\nI created a new project using the same code, but the init was successful.\r\nI cannot share the real project code. This project is a system-signed application and is an activity-less application. Will this have any impact?", "Hi @jahonn, Interesting ... a good data point for now, I suspect then that it is something incongruent with your gradle setups, can you maybe try to verify what the differences are with your new project gradle setup and your real project's gradle setup. If you are comfortable sharing your gradle files (perhaps you can censor things that you are confident are irrelevant) for both we can take a look.", "hi @pkgoogle ,\r\n\r\napp module's gradle file as below.\r\n`android {\r\n compileSdkVersion = 31\r\n buildToolsVersion = \"29.0.3\"\r\n\r\n defaultConfig {\r\n minSdkVersion 24\r\n targetSdkVersion 33\r\n applicationId 'xxxxxxxxxx'\r\n\r\n versionCode getVersionCode(defaultMajorVersionCode, defaultMinorVersionCode)\r\n versionName getVersionName(defaultMajorVersionName, defaultMinorVersionCode)\r\n testInstrumentationRunner 'androidx.test.runner.AndroidJUnitRunner'\r\n ndk {\r\n abiFilters \"armeabi\", \"armeabi-v7a\", \"arm64-v8a\", \"x86\"\r\n }\r\n\r\n // Enabling multiDex support.\r\n multiDexEnabled true\r\n }\r\n\r\n testOptions {\r\n unitTests.returnDefaultValues = true\r\n }\r\n\r\n compileOptions {\r\n sourceCompatibility JavaVersion.VERSION_1_8\r\n targetCompatibility JavaVersion.VERSION_1_8\r\n }\r\n\r\n sourceSets {\r\n main {\r\n manifest.srcFile 'AndroidManifest.xml'\r\n java.srcDirs = ['src', 'src_android', 'src_3rd']\r\n jni.srcDirs = ['jni']\r\n aidl.srcDirs = ['src', 'src_android', 'src_3rd']\r\n resources.srcDirs = ['src']\r\n renderscript.srcDirs = ['src']\r\n res.srcDirs = ['res']\r\n assets.srcDirs = ['assets']\r\n }\r\n\r\n // Move the tests to tests/java, tests/res, etc...\r\n androidTest.setRoot('androidTest')\r\n test.setRoot('test')\r\n\r\n // Move the build types to build-types/<type>\r\n // For instance, build-types/debug/java, build-types/debug/AndroidManifest.xml, ...\r\n // This moves them out of them default location under src/<type>/... which would\r\n // conflict with src/ being used by the main source set.\r\n // Adding new build types or product flavors should be accompanied\r\n // by a similar customization.\r\n debug.setRoot('build-types/debug')\r\n release.setRoot('build-types/release')\r\n }\r\n\r\n\r\n buildTypes {\r\n debug {\r\n testCoverageEnabled false\r\n minifyEnabled false\r\n proguardFiles getDefaultProguardFile('proguard-android.txt'), 'proguard-rules.pro'\r\n }\r\n release {\r\n testCoverageEnabled false\r\n minifyEnabled true\r\n shrinkResources true\r\n proguardFiles getDefaultProguardFile('proguard-android.txt'), 'proguard-rules.pro'\r\n }\r\n }\r\n\r\n repositories {\r\n flatDir {\r\n dirs 'libs'\r\n }\r\n }\r\n}\r\n`\r\n\r\nand root gradle files' classpath as below\r\n\r\n` classpath 'com.android.tools.build:gradle:4.2.2'\r\n classpath \"com.mobbeel.plugin:fat-aar:2.0.1\"\r\n classpath \"com.vanniktech:gradle-android-junit-jacoco-plugin:0.16.0\"\r\n classpath 'com.google.gms:google-services:4.3.3' // Google Services plugin\r\n classpath 'com.google.firebase:firebase-crashlytics-gradle:2.2.0'\r\n`", "Hi @jahonn, can you confirm you are fully following the directions on this page? There's a possibility that it has been updated since you last read it. https://www.tensorflow.org/lite/android/play_services#using_the_interpreter_apis\r\n\r\nI guess I meant the build.gradle files ... I was looking for the files which should include something like this: \r\n\r\n```kt\r\ndependencies {\r\n...\r\n // Tensorflow Lite dependencies for Google Play services\r\n implementation 'com.google.android.gms:play-services-tflite-java:16.0.1'\r\n // Optional: include Tensorflow Lite Support Library\r\n implementation 'com.google.android.gms:play-services-tflite-support:16.0.1'\r\n...\r\n}\r\n```", "Hi @pkgoogle \r\n\r\nI'm using this version\r\n‘\r\n // Tensorflow Lite dependencies\r\n implementation 'com.google.android.gms:play-services-tflite-java:16.1.0'\r\n\r\n’", "and the whole build.gradle file is \r\nplugins {\r\n id 'com.android.library'\r\n}\r\n\r\nandroid {\r\n namespace 'xxxxxxxx'\r\n compileSdkVersion rootProject.ext.compileSdkVersion\r\n\r\n defaultConfig {\r\n minSdkVersion rootProject.ext.minSdkVersion\r\n targetSdkVersion rootProject.ext.targetSdkVersion\r\n }\r\n\r\n buildTypes {\r\n release {\r\n minifyEnabled false\r\n proguardFiles getDefaultProguardFile('proguard-android-optimize.txt'), 'proguard-rules.pro'\r\n }\r\n }\r\n compileOptions {\r\n sourceCompatibility JavaVersion.VERSION_1_8\r\n targetCompatibility JavaVersion.VERSION_1_8\r\n }\r\n\r\n sourceSets {\r\n main {\r\n aidl.srcDirs += [\"src/main/java\"]\r\n }\r\n }\r\n}\r\n\r\ndependencies {\r\n // Tensorflow Lite dependencies\r\n implementation 'com.google.android.gms:play-services-tflite-java:16.1.0'\r\n ......\r\n}", "Hi @jahonn, Can you add this to your build.gradle dependencies:\r\n\r\n```java\r\nimplementation 'com.google.android.gms:play-services-tflite-support:16.0.1'\r\n```\r\n\r\nadditionally did you do these steps before using the Interpreter API? https://www.tensorflow.org/lite/android/play_services#using_the_interpreter_apis\r\n\r\n2. Add initialization of TensorFlow Lite:\r\n```java\r\nTask<Void> initializeTask = TfLite.initialize(context);\r\n```\r\n\r\n3. Ensure you have these dependencies\r\n```java\r\nimport org.tensorflow.lite.InterpreterApi\r\nimport org.tensorflow.lite.InterpreterApi.Options.TfLiteRuntime\r\n...\r\nprivate InterpreterApi interpreter;\r\n...\r\n```\r\n\r\nbefore the code you originally posted?", "Hi @pkgoogle \r\n\r\nI added the support lib, but still get the error.\r\n```\r\n implementation 'com.google.android.gms:play-services-tflite-java:16.0.1'\r\n implementation 'com.google.android.gms:play-services-tflite-support:16.0.1'\r\n\r\n```\r\n\r\ncode as below \r\n```\r\n Task<Void> initializeTask = TfLite.initialize(GlobalHolder.getApplicationContext());\r\n ByteBuffer buffer = ModelManager.getInstance().getModelBuffer(); // model from asserts\r\n\r\n initializeTask\r\n .addOnSuccessListener(a -> {\r\n MLog.i(TAG, \"load model success\");\r\n interpreter = InterpreterApi.create(buffer,\r\n new InterpreterApi.Options().setRuntime(TfLiteRuntime.FROM_SYSTEM_ONLY));\r\n })\r\n .addOnFailureListener(e -> {\r\n MLog.e(\"Interpreter\", String.format(\"Cannot initialize interpreter: %s\",\r\n e.getMessage()));\r\n });\r\n\r\n```\r\n\r\nerror as below\r\n\r\n`dlerror set to \"dlopen failed: couldn't map \"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\" segment 1: Permission denied\"`\r\n", "Hi @jahonn, are you using a real phone or an emulator? For both cases, can you ensure the Google Play Services app has as permissive permissions as possible, and then try running your projects again?", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "@pkgoogle hi, I using a real phone. All permissions the Google Play Services app need is allowed.", "Hi @jahonn, I think there is a permission difference between the two projects you tested, can you ensure your Google Play Services are updated and that your SDK version is above or equal to the targetSdkVersion? Additional, can you look at the logcat to see if there are permission errors prior to error message, also it may be worth it to compare logs from your new project that worked and your current project which doesn't work. Feel free to post it excluding PII for us to take a look.", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62407\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62407\">No</a>\n", "The relevant error messages seem to be these ones:\r\n\r\n```\r\n11-16 09:18:59.085 15532 15532 W TFLiteClient-0: type=1400 audit(0.0:2479): avc: denied { execute } for path=\"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk\" dev=\"dm-41\" ino=20657 scontext=u:r:platform_app:s0:c512,c768 tcontext=u:object_r:privapp_data_file:s0:c512,c768 tclass=file permissive=0 app=xxxxxxx\r\n```\r\n\r\nand\r\n\r\n```\r\n11-17 10:14:03.811 17066 17688 D linker : ... dlopen failed: couldn't map \"/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so\" segment 1: Permission denied\r\n```\r\n\r\nand\r\n\r\n```\r\n11-16 09:18:59.148 15532 15532 W System.err: android.os.RemoteException: Error loading TFLite native library\r\n11-16 09:18:59.148 15532 15532 W System.err: at qb.a(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:0)\r\n11-16 09:18:59.149 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.TfLiteDynamiteLoaderImpl.b(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:3)\r\n```\r\n\r\nBasically the issue is that the SELinux security policy in your phone's OS is denying your app execute permission to the \"TF Lite in Play services\" APK file\r\n`/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk`\r\nwhich contains the native code shared library file `lib/arm64-v8a/libtflite_gmscore_jni.so` embedded within it.\r\nThe file descriptor for that file is opened by the \"TF Lite in Play services\" module of Play services, and then passed back to your app, and then passed to `android_dlopen_ext` by the \"TF Lite in Play services\" client library.\r\n\r\nI suspect this is likely due to a misconfiguration of the SELinux policy for the device?\r\n\r\nIf my diagnosis is correct, a proper fix would require updating the SELinux policy configuration files that the OS provides.\r\n\r\nA likely work-around is to run the following command (requires root):\r\n\r\n```\r\nadb shell setenforce 0\r\n```\r\n\r\nIt may be worth raising this issue with the OEM that supplies the OS for your device.\r\nYour device was `Redmi/thunder_global/thunder:13/TP1A.220624.014/V0.0.0.0.XXXXXXX:user/test-keys`,\r\nso I guess that would be Redmi.\r\n\r\n", "@fergushenderson hi, thank you for your response. It's been a long time, and I also flashed a new ROM. I will rewrite a demo and try it on the new ROM.", "@fergushenderson \r\nThank you! I init the interpreter successfully with the command!" ]	2023-11-16T01:24:20	2024-04-22T03:00:11	2024-01-10T01:49:26	NONE	null	null	null	System information - Android Device information (use `adb shell getprop ro.build.fingerprint` if possible): Redmi/thunder_global/thunder:13/TP1A.220624.014/V0.0.0.0.XXXXXXX:user/test-keys - TensorFlow Lite in Play Services SDK version (found in `build.gradle`):com.google.android.gms:play-services-tflite-java:16.1.0 - Google Play Services version (`Settings` > `Apps` > `Google Play Services` > `App details`):23.43.13(190400-577232161) Standalone code to reproduce the issue ` Task<Void> initializeTask = TfLite.initialize(GlobalHolder.getApplicationContext()); ByteBuffer buffer = ModelManager.getInstance().getModelBuffer(); initializeTask .addOnSuccessListener(a -> { interpreter = InterpreterApi.create(buffer, new InterpreterApi.Options().setRuntime(InterpreterApi.Options.TfLiteRuntime.FROM_SYSTEM_ONLY)); MLog.i(TAG, "load model success"); }) .addOnFailureListener(e -> { e.printStackTrace(); MLog.e("Interpreter", String.format("Cannot initialize interpreter: %s", e.getMessage())); }); ` Any other info / logs 11-16 09:18:59.020 15532 16133 W DynamiteModule: Local module descriptor class for com.google.android.gms.tflite_dynamite not found. 11-16 09:18:59.046 15532 16133 I DynamiteModule: Considering local module com.google.android.gms.tflite_dynamite:0 and remote module com.google.android.gms.tflite_dynamite:231810005 11-16 09:18:59.046 15532 16133 I DynamiteModule: Selected remote version of com.google.android.gms.tflite_dynamite, version >= 231810005 11-16 09:18:59.047 15532 16133 V DynamiteModule: Dynamite loader version >= 2, using loadModule2NoCrashUtils 11-16 09:18:59.085 15532 15532 W TFLiteClient-0: type=1400 audit(0.0:2479): avc: denied { execute } for path="/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk" dev="dm-41" ino=20657 scontext=u:r:platform_app:s0:c512,c768 tcontext=u:object_r:privapp_data_file:s0:c512,c768 tclass=file permissive=0 app=xxxxxxx 11-16 09:18:59.147 15532 15716 I AD-PLUGIN-OnDeviceIntelligenceCore: OnDeviceIntelligence calculate failed, model load failed 11-16 09:18:59.148 15532 15532 W System.err: android.os.RemoteException: Error loading TFLite native library 11-16 09:18:59.148 15532 15532 W System.err: at qb.a(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:0) 11-16 09:18:59.149 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.TfLiteDynamiteLoaderImpl.b(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:3) 11-16 09:18:59.149 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.TfLiteDynamiteLoaderImpl.getInternalNativeInitializationHandleWithParams(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:0) 11-16 09:18:59.149 15532 15532 W System.err: at om.w(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:6) 11-16 09:18:59.149 15532 15532 W System.err: at bs.onTransact(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:4) 11-16 09:18:59.149 15532 15532 W System.err: at android.os.Binder.transact(Binder.java:1169) 11-16 09:18:59.149 15532 15532 W System.err: at com.google.android.gms.internal.tflite.zza.zzb(com.google.android.gms:play-services-tflite-impl@@16.1.0:2) 11-16 09:18:59.150 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.zza.zzf(com.google.android.gms:play-services-tflite-impl@@16.1.0:4) 11-16 09:18:59.150 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.internal.zzk.zzc(com.google.android.gms:play-services-tflite-impl@@16.1.0:11) 11-16 09:18:59.150 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.internal.zzi.zza(com.google.android.gms:play-services-tflite-impl@@16.1.0:9) 11-16 09:18:59.150 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.internal.zzf.then(Unknown Source:6) 11-16 09:18:59.150 15532 15532 W System.err: at com.google.android.gms.tasks.zzo.run(com.google.android.gms:play-services-tasks@@18.0.2:1) 11-16 09:18:59.151 15532 15532 W System.err: at java.util.concurrent.ThreadPoolExecutor.runWorker(ThreadPoolExecutor.java:1137) 11-16 09:18:59.151 15532 15532 W System.err: at java.util.concurrent.ThreadPoolExecutor$Worker.run(ThreadPoolExecutor.java:637) 11-16 09:18:59.151 15532 15532 W System.err: at java.lang.Thread.run(Thread.java:1012) 11-16 09:18:59.152 15532 15532 W System.err: Caused by: java.lang.UnsatisfiedLinkError: dlopen failed: couldn't map "/data/user_de/0/com.google.android.gms/app_chimera/m/00000016/dl-TfliteDynamiteDynamite.integ_231810005100400.apk!/lib/arm64-v8a/libtflite_gmscore_jni.so" segment 1: Permission denied 11-16 09:18:59.152 15532 15532 W System.err: at java.lang.Runtime.loadLibrary0(Runtime.java:1077) 11-16 09:18:59.152 15532 15532 W System.err: at java.lang.Runtime.loadLibrary0(Runtime.java:998) 11-16 09:18:59.152 15532 15532 W System.err: at java.lang.System.loadLibrary(System.java:1661) 11-16 09:18:59.152 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.TfLiteNativeApi.b(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:2) 11-16 09:18:59.153 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.TfLiteNativeApi.a(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:0) 11-16 09:18:59.153 15532 15532 W System.err: at com.google.android.gms.tflite.dynamite.TfLiteDynamiteLoaderImpl.b(:com.google.android.gms.policy_tflite_dynamite_dynamite@[email protected]:2) 11-16 09:18:59.153 15532 15532 W System.err: ... 13 more	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62407/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62407/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62406	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62406/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62406/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62406/events	https://github.com/tensorflow/tensorflow/issues/62406	1,995,745,772	I_kwDOArmXAs529Kns	62,406	Corrupt png files	{ "login": "gktval", "id": 7605046, "node_id": "MDQ6VXNlcjc2MDUwNDY=", "avatar_url": "https://avatars.githubusercontent.com/u/7605046?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gktval", "html_url": "https://github.com/gktval", "followers_url": "https://api.github.com/users/gktval/followers", "following_url": "https://api.github.com/users/gktval/following{/other_user}", "gists_url": "https://api.github.com/users/gktval/gists{/gist_id}", "starred_url": "https://api.github.com/users/gktval/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gktval/subscriptions", "organizations_url": "https://api.github.com/users/gktval/orgs", "repos_url": "https://api.github.com/users/gktval/repos", "events_url": "https://api.github.com/users/gktval/events{/privacy}", "received_events_url": "https://api.github.com/users/gktval/received_events", "type": "User", "site_admin": false }	[ { "id": 473184161, "node_id": "MDU6TGFiZWw0NzMxODQxNjE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:support", "name": "type:support", "color": "159b2e", "default": false, "description": "Support issues" } ]	closed	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "Running through all the images, I found the one causing the problem. For some reason or another it could be opened in other programs, but not in TF.\r\n```\r\nfor imgPath in images:\r\n img = tf.keras.utils.load_img(\r\n imgPath, color_mode=\"rgba\", target_size=(img_height, img_width)\r\n )\r\n\r\n img_array = tf.keras.utils.img_to_array(img)\r\n```", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62406\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62406\">No</a>\n" ]	2023-11-15T23:22:30	2023-11-15T23:46:17	2023-11-15T23:46:15	NONE	null	null	null	### Issue type Support ### TensorFlow version 2.14 ### OS platform and distribution Windows 11 ### Python version 3.9 I am getting an error: Invalid PNG data, size 49253 Part of my classification process classifying bad images from good ones. All of my png files can be opened, so I think the images are getting corrupt with they are resized. In the graphic below are a few of the "bad" images. ![image](https://github.com/tensorflow/tensorflow/assets/7605046/bd18cefb-f001-41a3-8f57-b14379cebdc4) After importing the images into the training dataset (code below), how I can filter out the images that became corrupt in the resizing? Right now it is throwing an error on when iterating through the normalization_layer ``` batch_size = 32 img_height = 224 img_width = 224 band_count = 4 train_ds = tf.keras.utils.image_dataset_from_directory( data_dir, validation_split=0.2, subset="training", seed=123, image_size=(img_height, img_width), batch_size=batch_size, color_mode="rgba") normalization_layer = layers.Rescaling(1./255) normalized_ds = train_ds.map(lambda x, y: (normalization_layer(x), y)) #Error here: Invalid PNG data, size 49253 image_batch, labels_batch = next(iter(normalized_ds)) ``` ### Standalone code to reproduce the issue ```shell Code above ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62406/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62406/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62405	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62405/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62405/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62405/events	https://github.com/tensorflow/tensorflow/pull/62405	1,994,468,041	PR_kwDOArmXAs5fgKdG	62,405	Prevent OSV scanner running in forks	{ "login": "elfringham", "id": 10442001, "node_id": "MDQ6VXNlcjEwNDQyMDAx", "avatar_url": "https://avatars.githubusercontent.com/u/10442001?v=4", "gravatar_id": "", "url": "https://api.github.com/users/elfringham", "html_url": "https://github.com/elfringham", "followers_url": "https://api.github.com/users/elfringham/followers", "following_url": "https://api.github.com/users/elfringham/following{/other_user}", "gists_url": "https://api.github.com/users/elfringham/gists{/gist_id}", "starred_url": "https://api.github.com/users/elfringham/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/elfringham/subscriptions", "organizations_url": "https://api.github.com/users/elfringham/orgs", "repos_url": "https://api.github.com/users/elfringham/repos", "events_url": "https://api.github.com/users/elfringham/events{/privacy}", "received_events_url": "https://api.github.com/users/elfringham/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169364259, "node_id": "MDU6TGFiZWwxMTY5MzY0MjU5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:XS", "name": "size:XS", "color": "adafea", "default": false, "description": "CL Change Size: Extra Small" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[]	2023-11-15T10:16:26	2023-12-04T09:48:35	2023-11-29T05:16:38	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62405", "html_url": "https://github.com/tensorflow/tensorflow/pull/62405", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62405.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62405.patch", "merged_at": "2023-11-29T05:16:38" }	The OSV scanner just fails anyway, so stop it running in forks to stop the failures being logged and emailed.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62405/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62405/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62404	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62404/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62404/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62404/events	https://github.com/tensorflow/tensorflow/issues/62404	1,994,174,456	I_kwDOArmXAs523K_4	62,404	I can't install tensorflow on termux on android	{ "login": "Noob225566", "id": 148705873, "node_id": "U_kgDOCN0SUQ", "avatar_url": "https://avatars.githubusercontent.com/u/148705873?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Noob225566", "html_url": "https://github.com/Noob225566", "followers_url": "https://api.github.com/users/Noob225566/followers", "following_url": "https://api.github.com/users/Noob225566/following{/other_user}", "gists_url": "https://api.github.com/users/Noob225566/gists{/gist_id}", "starred_url": "https://api.github.com/users/Noob225566/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Noob225566/subscriptions", "organizations_url": "https://api.github.com/users/Noob225566/orgs", "repos_url": "https://api.github.com/users/Noob225566/repos", "events_url": "https://api.github.com/users/Noob225566/events{/privacy}", "received_events_url": "https://api.github.com/users/Noob225566/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 2788658540, "node_id": "MDU6TGFiZWwyNzg4NjU4NTQw", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite-support", "name": "comp:lite-support", "color": "4A5F20", "default": false, "description": "Lite support library related issues" } ]	closed	false	{ "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false }	[ { "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false } ]	null	[ "@Noob225566 Could you please let us know if you have installed termux already and updated the packages?\r\nPlease let us know all the steps you followed which will help us to analyze this issue.\r\nThank you!", "you should do it in proot ubuntu/debian", "Sorry for me mistake that i type i can install tensorflow instead of i can't install tensorflow", "Sorry for these mistakes\r\nIssues type is not bug i can't install tensorflow in termux in android\r\n", "Just", "Sorry that i accidentally closed this issue", "@Noob225566 Please refer to the following steps for installation of TF on termux on android;\r\n\r\n1. Launch Termux.\r\n2. Make sure Termux is up to dated: pkg upgrade\r\n```\r\npkg upgrade\r\n```\r\n3. Install wget and proot: pkg install wget proot\r\n```\r\npkg install wget proot\r\n```\r\n4. Create directory for installing ubuntu: mkdir ubuntu && cd ubuntu\r\n```\r\nmkdir ubuntu && cd ubuntu\r\n```\r\n5. Download ubuntu chroot installation script: \r\n```\r\nwget https://raw.githubusercontent.com/Neo-Oli/termux-ubuntu/master/ubuntu.sh\r\nRun the installation script: bash ubuntu.sh\r\n```\r\n6. Now, we can start using Ubuntu by running the script: ./start-ubuntu.sh\r\n```\r\nchmod +x ubuntu.sh\r\n```\r\n```\r\nbash ubuntu.sh\r\n```\r\n\r\n7. Next, we will check and install updates on the Ubuntu image.\r\n\r\napt-get update && apt-get upgrade\r\n\r\n8. Install some prerequisites.\r\n```\r\napt-get install git wget vim zip build-essential python python3 python3-dev python3-pip libhdf5-dev\r\n```\r\n9. Install the TensorFlow package: To install the TensorFlow package, use the following command:\r\n```\r\npip install /path/to/built/tensorflow_package.whl\r\n```\r\nMake sure to replace /path/to/built/tensorflow_package.whl with the actual path to the built TensorFlow package.\r\n\r\n10. Verify TensorFlow installation: It is important to verify if the installation was successful. To do this, use the following command:\r\n```\r\npython -c \"import tensorflow as tf; print(tf.__version__)\"\r\n```\r\n\r\nPlease have a look at [this](https://medium.com/analytics-vidhya/developing-tensorflow-on-android-phone-cfc4297b676e) article as well. Hope it helps?\r\n\r\nThank you! ", "Thanks but it is a long process i need i small process i just want to make DL models in android i thought i can't use keras without tensorflow and keras is the most important labrairy for me to make DL models can i also use keras on android with tflite instead of tensorflow can you also help me in my one more issue which is i can't install h5py in termux ", "If you type my h5py issues in chrome you got my issus plz help me thanks", "Hi @Noob225566, \r\n\r\nYour Query : Can i also use keras on android with tflite instead of tensorflow.\r\nResponse: As per my knowledge you can convert your keras model to tflite and can run on android. \r\n\r\nYour Query: I can't install tensorflow in termux\r\n\r\nResponse: Please let us understand first, whether you want to install tensorflow or tensorflow lite on android. If you are using selected ops in your code then you need to install tensorflow build from source(Refer[ doc ](https://www.tensorflow.org/install/source)) or else if you have tflite file, install it from build source(Follow [doc](https://www.tensorflow.org/lite/guide/build_arm)) according to your android device.\r\n\r\n Please, let us know if it helps you.\r\n\r\nThank You", "Bro my only one question is can i use keras without tensorflow because every command i run using keras in termux i got tensorflow.comapt.v2 as tf no module found why can i fix this issue without installing tensorflow in termux i already installed tflite and theano and pytorch in termux i just wanna to use keras without tensorflow plz help me plz respond me as soon as possible also help me in my another issue that i suggested you plz respond me as soon as possible", "@pkgoogle,\r\nPlease look into the issue.\r\n\r\nThank You.", "What you mean by look into the issus i think no one ever read the issus completely because it is boring and hard to understand a computer issue that's why i only tell you the no module named tensorflow.compat.v2 as tf error and plz if you help me to install tensorflow in my low end android phone i really got satifies with you i have only one way to install tensorflow on android which is cloning the tensorflow library using git clone but it takes too much time and data and storage because tensorflow is a too big Library also i think it some chances of errors and couldn't work properly so plz help me to install tensorflow easily or tell me any other way to use keras without tensorflow plz respond me as soon as possible", "Hi @Noob225566, that previous comment was directed at me :).\r\n\r\nAs I understand it, you want to try to use Keras independent of TensorFlow? because TF is too big for your low end Android device?\r\n\r\nkeras is now available in multiple backends (that is you can switch the backend to either JAX or Pytorch) now: https://keras.io/. However 2 GB of RAM seems too small even for those alternatives. May I ask if you really want to use Keras on the android? Are you using it as an ML development machine? if so you may have better luck starting with colab: https://colab.sandbox.google.com/. Generally speaking you probably shouldn't use a mobile/edge device as a development machine but more as an inference device, which is what TF Lite is specialized for. See if you either of those 2 solutions work better for you. If they don't meet your requirements let us know and we'll see how we can help.", "Thanks to answer my question that i can even use keras even without installing tensorflow i have this doubt because an AI bard said me this that i need tensorflow is necessary to use keras or you can't even change its backend as you ask me if i really want to use keras on android the answer is yes i am in a middle class family in india and i am 11 years old i have a vast level of interest in programming,ML,DL and AI that's why i wanna to use keras in my mother's low end device android phone i already know programming and ML i just wanna to create small models like spam filter,image classifier and chatbots some more small or medium DL models in my mother's android phone but i have another doubt as you told me i can change keras backend but i try every possible way to change keras backend using termux (where i installed keras) but i can't do that then i see keras-core that and then which claimed it is a multi backend version of keras but i can't install keras-core in termux in android because h5py is not installed and whenever i run h5py i got a cython compile error which i can't fix can you help me plz respond me as soon as possible plz", "Hi @Noob225566, have you tried using keras on colab? https://colab.sandbox.google.com/ . You may want to raise an issue on the keras.io github repo: https://github.com/keras-team/keras-io, they may be able to better support your issues with changing the backend.", "Bro thanks to suggest me to use colab when i run this command\r\nfrom keras.layers import Dense\r\nprint (\"it works\") \r\nIt shows it works means there is no module error thanks for this but this is online and i wanna offline because i don't like online things well do you know about a python interpreter which name is \"pydroid 3\" it's my favourite python IDE because it's GUI is very simple,we can run programs in one click and it's offline but tensorflow and pytorch in pydroid 3 are for only premium which i can't afford that's why i decided to use termux but it also have issues i think colab is a good choice but can you tell me any other python IDE like pydroid 3 for Android which is similar like pydroid 3 and offline and supports keras and tensorflow for free plz respond me as soon as possible thanks", "Hi @Noob225566, libraries/packages (such as TF or PT) should be independent of IDEs. Termux isn't really an IDE but more of a terminal emulator on android, for IDEs in a low resource environment ... the classics like vi(m) or emacs may be more viable, I think neovim might be the better balance for features but still being low resource. That being said you still need TF working with Termux... what error are you receiving there? or do you already know that TF is too big for your system? If that's the case then IDE won't matter, which is why I recommended Colab as you get resources from Google essentially for free. I understand it may not be the ideal environment for you but I'm trying my best to help you with what you have. Is there something you feel you can't do on colab that you want to do on android/termux?", "Thanks for your early response you are right i already setup termux with neovim so i can get a good GUI like pydroid 3 and it works every package or library i install on termux will also works in neovim in termux that's why if i download TF in termux i can use it in neovim in termux but i know TF is too big for my low end android device well as you ask me the error i got so i wanna tell you that whenever i try to run keras in termux or neovim i got this error (no module found tensorflow.comapt.v2 as tf) this is just a small amount of error if you ask me i can provide you full error also i can't install tensorflow using pip when i try to install TF with pip i got no module found that satisfies the requirement tensorflow that's why but now i think colab is a good choice thanks for your suggestion but the only thing i don't like about is that it is online but i wanna offline things but ok it is also helpful i wish you reply to my this chat as soon as possible now i am going to make my DL model in colab", "Can you try on android/termux:\r\n\r\n```\r\nsudo apt-get install git wget vim zip build-essential python python3 python3-dev python3-pip libhdf5-dev\r\npip install --upgrade pip\r\npip install tensorflow\r\n```\r\n\r\nLet me know what errors you get", "When i run these commands in termux in android i got these errors\r\nERROR: Installing pip is forbidden, this will break the python-pip package (termux).\r\nDEPRECATION: Loading egg at /data/data/com.termux/files/usr/lib/python3.11/site-packages/dm_tree-0.1.8-py3.11-linux-armv7l.egg is deprecated. pip 24.3 will enforce this behaviour change. A possible replacement is to use pip for package installation.. Discussion can be found at https://github.com/pypa/pip/issues/12330\r\nERROR: Could not find a version that satisfies the requirement tensorflow (from versions: none)\r\nERROR: No matching distribution found for tensorflow\r\nAlso plz tell me i am successfull to download tensorflow from pip does it takes too much storage like even i download tensorflow from pip does it takes 700-800mb if yes so i don't wanna download tensorflow on termux plz respond me as soon as possible", "Hi @Noob225566, your workflow isn't really well supported. I am fairly sure the package is quite large overall maybe around the size you mention. I think the best thing for you would be to just use colab. It seems pip is having trouble finding a version of TF that works with your hardware. Is there any other message after this:\r\n\r\n```\r\nERROR: Could not find a version that satisfies the requirement tensorflow (from versions: none)\r\nERROR: No matching distribution found for tensorflow\r\n```\r\n\r\nYou can attempt manually installing earlier versions of TF to see if any of them work (ex):\r\n```\r\npip install tensorflow==2.15\r\n```\r\n\r\nMaybe try until 2.10 and see if any of them work. If not I don't believe we have a version which would work with your system.", "Bro i just wanna tensorflow to use keras because i do everything to change keras backend even i installed keras-core (multi backend version of keras) but nothing works and always say me no module named tensorflow.comapt.v2 as tf this is the problem that i wanna to fix i already have theano and pytorch installed but i can't change its backend 😭😭😭 plz help me as soon as possible", "Bro plz response me as soon as possible", "Hi @Noob225566, I don't believe TensorFlow is available for your system, please communicate with https://github.com/keras-team/keras-io to see how you may change the backend. Apologies for not being able to help further. ", "Yeah brother i already make a issues in keras repository i got my first respond but didn't work it's ok if you can't help me anymore if you know any interesting knowledge on my issue plz tell me also thanks for your replies and suggestion of colab thanks goodbye", "Hi @Noob225566, if there are no more open items for this issue, can you please close the issue as \"Close as not planned\" if you have no more open items? Thanks.", "Ok i will now closing the issus thanks for the help of you thanks" ]	2023-11-15T07:05:17	2023-12-06T09:32:58	2023-12-06T09:32:58	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version I even don't install it ### Custom code Yes ### OS platform and distribution Idk ### Mobile device Redmi 7A android device with 2gb ram and 32gb storage ### Python version I think higher than 3.8 i think 3.11 ### Bazel version I can provide you later ### GCC/compiler version I can provide you later ### CUDA/cuDNN version I can provide you later ### GPU model and memory I can provide you later ### Current behavior? I can install tensorflow or tensorflow lite in terms in android plz help me if you need and information just ask me and i provide you the information plz help me ### Standalone code to reproduce the issue ```shell pip install tensorflow ``` ### Relevant log output ```shell Just ask me and i wil provide you ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62404/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62404/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62403	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62403/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62403/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62403/events	https://github.com/tensorflow/tensorflow/issues/62403	1,994,141,504	I_kwDOArmXAs523C9A	62,403	How to get c++ code coverage when running Python tests?	{ "login": "BiophiliaSWDA", "id": 127826463, "node_id": "U_kgDOB556Hw", "avatar_url": "https://avatars.githubusercontent.com/u/127826463?v=4", "gravatar_id": "", "url": "https://api.github.com/users/BiophiliaSWDA", "html_url": "https://github.com/BiophiliaSWDA", "followers_url": "https://api.github.com/users/BiophiliaSWDA/followers", "following_url": "https://api.github.com/users/BiophiliaSWDA/following{/other_user}", "gists_url": "https://api.github.com/users/BiophiliaSWDA/gists{/gist_id}", "starred_url": "https://api.github.com/users/BiophiliaSWDA/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/BiophiliaSWDA/subscriptions", "organizations_url": "https://api.github.com/users/BiophiliaSWDA/orgs", "repos_url": "https://api.github.com/users/BiophiliaSWDA/repos", "events_url": "https://api.github.com/users/BiophiliaSWDA/events{/privacy}", "received_events_url": "https://api.github.com/users/BiophiliaSWDA/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "Hello, I'm also struggling with this issue. I'm wondering if you've found a solution?" ]	2023-11-15T06:38:21	2024-02-26T05:11:38	null	NONE	null	null	null	It seems connect with these two issues([51091](https://github.com/tensorflow/tensorflow/issues/51091), [46477](https://github.com/tensorflow/tensorflow/issues/46477)), but they were all closed by robot. I try ``` bazel coverage -s --instrument_test_targets --coverage_report_generator=@bazel_tools//tools/test:coverage_report_generator --coverage_support=@bazel_tools//tools/test:coverage_support --collect_code_coverage --jobs 5 //tensorflow ``` and ``` sudo bazel coverage -s --instrument_test_targets --experimental_cc_coverage --coverage_report_generator=@bazel_tools//tools/test/CoverageOutputGenerator/java/com/google/devtools/coverageoutputgenerator:Main --combined_report=lcov --jobs 4 //tensorflow/core:all ``` The `.dat` files generated are all 0 Bytes. I want to use `bazel coverage` to calculate the `c++ code coverage`, when a python program runs. I don't know if this is possible. If not, can you provide a suitable way for me to obtain the `c++ code coverage`? My envs: ``` Ubuntu20.04 Tensorflow2.12.0 (Building from Source) bazel5.3.0 (using bazelisk) Python3.9 CUDA11.7 cuDnn8.5.0 GCC9.4.0 ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62403/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62403/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62402	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62402/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62402/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62402/events	https://github.com/tensorflow/tensorflow/pull/62402	1,994,128,285	PR_kwDOArmXAs5ffArk	62,402	Updated the description of max_delta parameter	{ "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 1169364259, "node_id": "MDU6TGFiZWwxMTY5MzY0MjU5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:XS", "name": "size:XS", "color": "adafea", "default": false, "description": "CL Change Size: Extra Small" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "I don't see \"absolute\" anywhere in the documentation. Instead, the docs refer to the `adjust_brightness` documentation, which would probably be the place to change if anything.", "Hi @sushreebarsa Can you please check @cantonios's [comments](https://github.com/tensorflow/tensorflow/pull/62402#issuecomment-1813075282) and keep us posted ? Thank you!\r\n\r\n\r\n\r\n" ]	2023-11-15T06:25:17	2023-11-28T16:43:52	2023-11-28T16:43:51	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62402", "html_url": "https://github.com/tensorflow/tensorflow/pull/62402", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62402.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62402.patch", "merged_at": "2023-11-28T16:43:51" }	The documentation currently states that the max_delta parameter controls the maximum absolute change in brightness of images. However, the max_delta parameter actually controls the maximum relative change in brightness of images. This means that the actual change in brightness will depend on the range of values in the input image. So, I have updated the description for the max_delta parameter. Could you please have a look and do the needful? Thank you!	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62402/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62402/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62401	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62401/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62401/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62401/events	https://github.com/tensorflow/tensorflow/pull/62401	1,993,858,519	PR_kwDOArmXAs5feGrM	62,401	[TOSA] Remove lower_global_tensors pass	{ "login": "Jerry-Ge", "id": 24451859, "node_id": "MDQ6VXNlcjI0NDUxODU5", "avatar_url": "https://avatars.githubusercontent.com/u/24451859?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Jerry-Ge", "html_url": "https://github.com/Jerry-Ge", "followers_url": "https://api.github.com/users/Jerry-Ge/followers", "following_url": "https://api.github.com/users/Jerry-Ge/following{/other_user}", "gists_url": "https://api.github.com/users/Jerry-Ge/gists{/gist_id}", "starred_url": "https://api.github.com/users/Jerry-Ge/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Jerry-Ge/subscriptions", "organizations_url": "https://api.github.com/users/Jerry-Ge/orgs", "repos_url": "https://api.github.com/users/Jerry-Ge/repos", "events_url": "https://api.github.com/users/Jerry-Ge/events{/privacy}", "received_events_url": "https://api.github.com/users/Jerry-Ge/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169365682, "node_id": "MDU6TGFiZWwxMTY5MzY1Njgy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:L", "name": "size:L", "color": "adafea", "default": false, "description": "CL Change Size: Large" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[]	2023-11-15T01:33:06	2023-12-11T19:14:26	2023-12-11T19:14:26	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62401", "html_url": "https://github.com/tensorflow/tensorflow/pull/62401", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62401.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62401.patch", "merged_at": "2023-12-11T19:14:26" }	- Removed lower_global_tensors pass which lowers TFL Variable operators to ML_Program Global operators - Instead, we should target TFL->TOSA->ML_Program pass	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62401/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62401/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62400	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62400/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62400/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62400/events	https://github.com/tensorflow/tensorflow/issues/62400	1,993,775,069	I_kwDOArmXAs521pfd	62,400	`log_softmax` could be `2102` to `2970` times more accurate	{ "login": "JasonGross", "id": 396076, "node_id": "MDQ6VXNlcjM5NjA3Ng==", "avatar_url": "https://avatars.githubusercontent.com/u/396076?v=4", "gravatar_id": "", "url": "https://api.github.com/users/JasonGross", "html_url": "https://github.com/JasonGross", "followers_url": "https://api.github.com/users/JasonGross/followers", "following_url": "https://api.github.com/users/JasonGross/following{/other_user}", "gists_url": "https://api.github.com/users/JasonGross/gists{/gist_id}", "starred_url": "https://api.github.com/users/JasonGross/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/JasonGross/subscriptions", "organizations_url": "https://api.github.com/users/JasonGross/orgs", "repos_url": "https://api.github.com/users/JasonGross/repos", "events_url": "https://api.github.com/users/JasonGross/events{/privacy}", "received_events_url": "https://api.github.com/users/JasonGross/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473173272, "node_id": "MDU6TGFiZWw0NzMxNzMyNzI=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:feature", "name": "type:feature", "color": "159b2e", "default": false, "description": "Feature requests" }, { "id": 1097545817, "node_id": "MDU6TGFiZWwxMDk3NTQ1ODE3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:apis", "name": "comp:apis", "color": "0052cc", "default": false, "description": "Highlevel API related issues" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "@sachinprasadhs I was able to replicate the issue reported [here](https://colab.research.google.com/gist/sushreebarsa/00223aae8b2afeab0da98122da85a040/62400.ipynb). Please find the attached gist. Thank you!" ]	2023-11-14T23:52:13	2023-12-06T00:19:45	null	NONE	null	null	null	### Issue type Feature Request ### Have you reproduced the bug with TensorFlow Nightly? No ### Source binary ### TensorFlow version unknown 2.12.0 ### Custom code No ### OS platform and distribution Linux 5.15.0-67-generic #74~20.04.1-Ubuntu SMP Wed Feb 22 14:52:34 UTC 2023 x86_64 GNU/Linux ### Mobile device _No response_ ### Python version 3.11.4 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? As I understand it, TensorFlow implements `log_softmax(x)` as `x - max(x) - log(sum(exp(x - max(x))))`. However, when the largest value is much larger than the rest of the values (about 16 larger for float32, about 36 larger for float64), `log_softmax` returns `0` at the maximum value, when it could give a much more precise answer by using `log1p` and zeroing out the maximum cell (guaranteed to be 1) before summing. This came up when a transformer I was training with cross-entropy loss on a classification task had loss dominated by float32 error. Might help with https://github.com/tensorflow/tensorflow/issues/48816, https://github.com/tensorflow/tensorflow/issues/12002 I originally posted this as a [StackOverflow question](https://stackoverflow.com/q/77477327/377022). Companion PyTorch issue: https://github.com/pytorch/pytorch/issues/113708 Companion SciPy issue: https://github.com/scipy/scipy/issues/19521 ### Standalone code to reproduce the issue ```python import tensorflow as tf eps = tf.constant(tf.keras.backend.epsilon(), dtype=tf.float32) print(-tf.nn.log_softmax(tf.constant([1 - tf.math.log(2 * eps).numpy(), 0], dtype=tf.float32))) # tf.Tensor([1.1920928e-07 1.6424950e+01], shape=(2,), dtype=float32) print(-tf.nn.log_softmax(tf.constant([1 - tf.math.log(eps).numpy(), 0], dtype=tf.float32))) # tf.Tensor([-0. 17.118095], shape=(2,), dtype=float32) ``` ```python import tensorflow as tf import numpy as np def log_softmax_alt(x): maxi = tf.argmax(x) xoffset = x - x[maxi] xoffsetexp = tf.exp(xoffset) # xoffsetexp[maxi] is currently about 1 xoffsetexp = tf.tensor_scatter_nd_update(xoffsetexp, [[maxi]], [0]) xoffsetexp_sum_m1 = tf.reduce_sum(xoffsetexp) return xoffset - tf.math.log1p(xoffsetexp_sum_m1) for ty in (tf.float32, tf.float64): smallest_log_softmax, smallest_log_softmax_alt, smallest_log_softmax_val, smallest_log_softmax_alt_val = 0, 0, 0, 0 # work around https://github.com/tensorflow/tensorflow/issues/62399 for i in range(int(1 - np.log2(tf.experimental.numpy.finfo(ty).smallest_subnormal))): values = tf.constant([1 + i, 0], dtype=ty) log_softmax_values = tf.nn.log_softmax(values) log_softmax_values_alt = log_softmax_alt(values) if log_softmax_values[0] != 0: smallest_log_softmax, smallest_log_softmax_val = i, log_softmax_values[0] if log_softmax_values_alt[0] != 0: smallest_log_softmax_alt, smallest_log_softmax_alt_val = i, log_softmax_values_alt[0] if log_softmax_values[0] == 0 and log_softmax_values_alt[0] == 0: break print(f"For {ty}, diff in supported input accuracy is 2-({smallest_log_softmax_alt} - {smallest_log_softmax}) = 2-{smallest_log_softmax_alt - smallest_log_softmax}; diff in output accuracy is np.log2({-smallest_log_softmax_val}) - np.log2({-smallest_log_softmax_alt_val}) = {np.log2(-smallest_log_softmax_val) - np.log2(-smallest_log_softmax_alt_val)}") ``` gives ``` For <dtype: 'float32'>, diff in supported input accuracy is 2-(86 - 15) = 2-71; diff in output accuracy is np.log2(1.1920928244535389e-07) - np.log2(1.6458114537543937e-38) = 102.51446533203125 For <dtype: 'float64'>, diff in supported input accuracy is 2-(707 - 35) = 2-672; diff in output accuracy is np.log2(2.2204460492503128e-16) - np.log2(3.3075530036384083e-308) = 969.428088949386 ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62400/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62400/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62399	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62399/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62399/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62399/events	https://github.com/tensorflow/tensorflow/issues/62399	1,993,768,243	I_kwDOArmXAs521n0z	62,399	`tf.math.log` thinks that -103 is -inf	{ "login": "JasonGross", "id": 396076, "node_id": "MDQ6VXNlcjM5NjA3Ng==", "avatar_url": "https://avatars.githubusercontent.com/u/396076?v=4", "gravatar_id": "", "url": "https://api.github.com/users/JasonGross", "html_url": "https://github.com/JasonGross", "followers_url": "https://api.github.com/users/JasonGross/followers", "following_url": "https://api.github.com/users/JasonGross/following{/other_user}", "gists_url": "https://api.github.com/users/JasonGross/gists{/gist_id}", "starred_url": "https://api.github.com/users/JasonGross/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/JasonGross/subscriptions", "organizations_url": "https://api.github.com/users/JasonGross/orgs", "repos_url": "https://api.github.com/users/JasonGross/repos", "events_url": "https://api.github.com/users/JasonGross/events{/privacy}", "received_events_url": "https://api.github.com/users/JasonGross/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 1478826728, "node_id": "MDU6TGFiZWwxNDc4ODI2NzI4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:core", "name": "comp:core", "color": "024391", "default": false, "description": "issues related to core part of tensorflow" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	open	false	{ "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false }	[ { "login": "SuryanarayanaY", "id": 116063290, "node_id": "U_kgDOBur8Og", "avatar_url": "https://avatars.githubusercontent.com/u/116063290?v=4", "gravatar_id": "", "url": "https://api.github.com/users/SuryanarayanaY", "html_url": "https://github.com/SuryanarayanaY", "followers_url": "https://api.github.com/users/SuryanarayanaY/followers", "following_url": "https://api.github.com/users/SuryanarayanaY/following{/other_user}", "gists_url": "https://api.github.com/users/SuryanarayanaY/gists{/gist_id}", "starred_url": "https://api.github.com/users/SuryanarayanaY/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/SuryanarayanaY/subscriptions", "organizations_url": "https://api.github.com/users/SuryanarayanaY/orgs", "repos_url": "https://api.github.com/users/SuryanarayanaY/repos", "events_url": "https://api.github.com/users/SuryanarayanaY/events{/privacy}", "received_events_url": "https://api.github.com/users/SuryanarayanaY/received_events", "type": "User", "site_admin": false } ]	null	[ "I believe to solve your issue you can try:\r\n\r\n`import tensorflow as tf\r\nimport numpy as np\r\n\r\neps_tf = tf.experimental.numpy.finfo(tf.float32).tiny\r\neps_np = np.finfo(np.float32).tiny\r\n\r\nprint(tf.math.log(eps_tf))\r\nprint(np.log(eps_np))\r\nprint(type(np.log(eps_np)))\r\n`", "@AymenS02 I specifically need the smallest subnormal float, not the smallest normal float, for testing the precision limits of `log_softmax`. I can solve my issue by using `np.log` instead of `tf.math.log`, this is not blocking my development; I just figured tf might want to fix this bug.", "Hi @JasonGross ,\r\n\r\nI have replicated the reported behaviour and attaching [gist](https://colab.sandbox.google.com/gist/SuryanarayanaY/bd8b1206037bcf2fd64d22daa84f0299/62399.ipynb) for reference. \r\n\r\nWill escalate to concerned SME. Thanks!", "@JasonGross TF (and XLA in general) intentionally flushes subnormals to zero. We do so at the start of every compute thread (on platforms that support such flushing) [[link](https://github.com/google/tsl/blob/a7362a4151095ad8cf8f9b65d1225fda62d11194/tsl/platform/threadpool.cc#L65)]. This significantly speeds up computations in most ML applications. So it's somewhat intentional, and not something we're going to \"fix\".", "Ah, I see. Makes sense, thanks for the explanation" ]	2023-11-14T23:45:40	2023-11-16T06:39:38	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source binary ### TensorFlow version unknown 2.12.0 ### Custom code No ### OS platform and distribution Linux 5.15.0-67-generic #74~20.04.1-Ubuntu SMP Wed Feb 22 14:52:34 UTC 2023 x86_64 GNU/Linux ### Mobile device _No response_ ### Python version 3.11.4 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? TensorFlow can't seem to handle taking the log of tiny subnormal floats, even though numpy can. (Sorry if this is already fixed, I haven't tested this on nightly.) ### Standalone code to reproduce the issue ```python import tensorflow as tf import numpy as np eps = tf.experimental.numpy.finfo(tf.float32).smallest_subnormal print(tf.math.log(eps)) print(np.log(eps)) print(type(np.log(eps))) ``` gives ``` tf.Tensor(-inf, shape=(), dtype=float32) -103.27893 <class 'numpy.float32'> ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62399/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62399/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62398	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62398/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62398/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62398/events	https://github.com/tensorflow/tensorflow/issues/62398	1,993,615,602	I_kwDOArmXAs521Cjy	62,398	tf.keras.__version__ is undefined since TF 2.14.0	{ "login": "ageron", "id": 76661, "node_id": "MDQ6VXNlcjc2NjYx", "avatar_url": "https://avatars.githubusercontent.com/u/76661?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ageron", "html_url": "https://github.com/ageron", "followers_url": "https://api.github.com/users/ageron/followers", "following_url": "https://api.github.com/users/ageron/following{/other_user}", "gists_url": "https://api.github.com/users/ageron/gists{/gist_id}", "starred_url": "https://api.github.com/users/ageron/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ageron/subscriptions", "organizations_url": "https://api.github.com/users/ageron/orgs", "repos_url": "https://api.github.com/users/ageron/repos", "events_url": "https://api.github.com/users/ageron/events{/privacy}", "received_events_url": "https://api.github.com/users/ageron/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1097545817, "node_id": "MDU6TGFiZWwxMDk3NTQ1ODE3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:apis", "name": "comp:apis", "color": "0052cc", "default": false, "description": "Highlevel API related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false }	[ { "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false } ]	null	[ "Hello, @ageron! \r\nIf you are using the latest version of TF then please use the following code;\r\n\r\nTo print the Keras version;\r\n```\r\nimport keras\r\n\r\nprint(keras.__version__)\r\n\r\n```\r\nTo print the TF version;\r\n\r\n```\r\nimport tensorflow as tf\r\n\r\nprint(tf.__version__)\r\n\r\n```\r\nPlease have a look at this [gist](https://colab.research.google.com/gist/sushreebarsa/c724800e284e5bc15d03fd6f25c9300e/62398.ipynb) for reference!\r\nThank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62398\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62398\">No</a>\n", "On 2.15.0, the same error still persists with `keras.__version__`\r\n\r\n```\r\n---------------------------------------------------------------------------\r\nAttributeError Traceback (most recent call last)\r\nCell In[8], [line 4](vscode-notebook-cell:?execution_count=8&line=4)\r\n [1](vscode-notebook-cell:?execution_count=8&line=1) import tensorflow as tf\r\n [2](vscode-notebook-cell:?execution_count=8&line=2) from tensorflow import keras\r\n----> [4](vscode-notebook-cell:?execution_count=8&line=4) tf.__version__, keras.__version__\r\n\r\nAttributeError: module 'keras.api._v2.keras' has no attribute '__version__'\r\n```", "Keras moved from being a library of TF (so import TF and you get Keras) to being a library that works with TF (and JAX, and PyTorch) as backend (so you import _both_ Keras and TF).\r\n\r\nTo not break existing code, some `tf.keras` symbols are still in place, but not all of them" ]	2023-11-14T21:38:18	2024-01-10T17:02:55	2023-12-01T01:52:05	CONTRIBUTOR	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version v2.14.0-rc1-21-g4dacf3f368e 2.14.0 ### Custom code No ### OS platform and distribution _No response_ ### Mobile device N/A ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? `tf.keras.__version__` is undefined. It was defined since TensorFlow 1.9, and until 2.13, but for some reason it looks like it was removed in TF 2.14. This change was not listed in the release notes. Not sure `tf.keras.__version__` was part of the official public API, but a quick search on the Web and github shows that it was heavily used. ### Standalone code to reproduce the issue ```shell import tensorflow as tf print(tf.keras.__version__) ``` ### Relevant log output ```shell In [2]: tf.keras.__version__ --------------------------------------------------------------------------- AttributeError Traceback (most recent call last) Cell In[2], line 1 ----> 1 tf.keras.__version__ File ~/opt/miniconda3/envs/homl3/lib/python3.10/site-packages/tensorflow/python/util/lazy_loader.py:67, in LazyLoader.__getattr__(self, item) 65 def __getattr__(self, item): 66 module = self._load() ---> 67 return getattr(module, item) AttributeError: module 'keras.api._v2.keras' has no attribute '__version__' ``` ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62398/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62398/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62397	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62397/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62397/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62397/events	https://github.com/tensorflow/tensorflow/issues/62397	1,993,467,411	I_kwDOArmXAs520eYT	62,397	Tensorflow terribly slow on Mac Studio M1 Ultra - problem with tensorflow-metal	{ "login": "francescosgarlata", "id": 128385439, "node_id": "U_kgDOB6cBnw", "avatar_url": "https://avatars.githubusercontent.com/u/128385439?v=4", "gravatar_id": "", "url": "https://api.github.com/users/francescosgarlata", "html_url": "https://github.com/francescosgarlata", "followers_url": "https://api.github.com/users/francescosgarlata/followers", "following_url": "https://api.github.com/users/francescosgarlata/following{/other_user}", "gists_url": "https://api.github.com/users/francescosgarlata/gists{/gist_id}", "starred_url": "https://api.github.com/users/francescosgarlata/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/francescosgarlata/subscriptions", "organizations_url": "https://api.github.com/users/francescosgarlata/orgs", "repos_url": "https://api.github.com/users/francescosgarlata/repos", "events_url": "https://api.github.com/users/francescosgarlata/events{/privacy}", "received_events_url": "https://api.github.com/users/francescosgarlata/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1205765054, "node_id": "MDU6TGFiZWwxMjA1NzY1MDU0", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:macOS", "name": "subtype:macOS", "color": "b619ea", "default": false, "description": "macOS Build/Installation issues" }, { "id": 1463677878, "node_id": "MDU6TGFiZWwxNDYzNjc3ODc4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:performance", "name": "type:performance", "color": "159b2e", "default": false, "description": "Performance Issue" }, { "id": 5508003926, "node_id": "LA_kwDOArmXAs8AAAABSE14Vg", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.13", "name": "TF 2.13", "color": "B13ACB", "default": false, "description": "For issues related to Tensorflow 2.13" } ]	closed	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @francescosgarlata ,\r\n\r\nI have replicated the issue. This problem observed with new optimizers earlier and now legacy optimizers also seems same problem. Since tensorflow-macos was built and maintained by Apple, could you please report the issue at [apple](https://developer.apple.com/forums/tags/tensorflow-metal/) dev forum ?\r\n\r\nAlso since this is keras related you can post this issue at [tf-keras](https://github.com/keras-team/tf-keras) repo also to track it by Keras team. \r\n\r\nThanks!", "Hi just in case it is helpful, it should be a M1-Ultra-specific issue. I cannot reproduce it on M1 Pro.", "@francescosgarlata , tensorflow-metal is not released by Google, we only support official CPU support for M1 when installed through `pip install tensorflow`.\r\n\r\nCould you please install tensorflow through pip and then install tensorflow-metal and see if that changes the performance.\r\nIf you notice the difference again, report the issue at https://developer.apple.com/forums/tags/tensorflow-metal/", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62397\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62397\">No</a>\n", "> Thank\r\n\r\nHi @sachinprasadhs and @francescosgarlata , we have seen this performance issue with moving to new Optimizer operations in Keras which are implemented \r\nhttps://github.com/tensorflow/tensorflow/issues/59438#issuecomment-1411414605\r\n" ]	2023-11-14T19:58:41	2024-02-21T10:10:44	2023-12-21T01:48:43	NONE	null	null	null	### Issue type Performance ### Have you reproduced the bug with TensorFlow Nightly? No ### Source binary ### TensorFlow version v2.13.0-rc2-7-g1cb1a030a62 2.13.0 ### Custom code No ### OS platform and distribution Mac OS 13.5.2 ### Mobile device _No response_ ### Python version 3.11 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I have a Mac Studio M1 Ultra and I am trying to train a simple RNN to forecast time series. When I run the code, tensorflow takes ~140s per epoch or 670ms/step. If I uninstall tensorflow-metal, it takes only 20ms/step. But in this way I won't be able to use the GPU. Why is this happening? ### Standalone code to reproduce the issue ```shell import numpy as np import keras import tensorflow as tf def generate_time_series(batch_size, n_steps): freq1, freq2, offsets1, offsets2 = np.random.rand(4, batch_size, 1) time = np.linspace(0, 1, n_steps) series = 0.5 * np.sin((time - offsets1) * (freq1 * 10 + 10)) # wave 1 series += 0.2 * np.sin((time - offsets2) * (freq2 * 20 + 20)) # + wave 2 series += 0.1 * (np.random.rand(batch_size, n_steps) - 0.5) # + noise return series[..., np.newaxis].astype(np.float32) np.random.seed(42) n_steps = 50 series = generate_time_series(10000, n_steps + 1) X_train, y_train = series[:7000, :n_steps], series[:7000, -1] X_valid, y_valid = series[7000:9000, :n_steps], series[7000:9000, -1] X_test, y_test = series[9000:, :n_steps], series[9000:, -1] # Implementing simple RRN np.random.seed(42) tf.random.set_seed(42) model = keras.models.Sequential([keras.layers.SimpleRNN(1, input_shape=[None, 1])]) optimizer = tf.keras.optimizers.legacy.Adam(learning_rate=0.005) model.compile(loss="mse", optimizer=optimizer) history = model.fit(X_train, y_train, epochs=20, validation_data=(X_valid, y_valid)) ``` ### Relevant log output ```shell Epoch 1/20 2023-11-14 20:58:05.842329: I metal_plugin/src/device/metal_device.cc:1154] Metal device set to: Apple M1 Ultra 2023-11-14 20:58:05.842356: I metal_plugin/src/device/metal_device.cc:296] systemMemory: 64.00 GB 2023-11-14 20:58:05.842361: I metal_plugin/src/device/metal_device.cc:313] maxCacheSize: 24.00 GB 2023-11-14 20:58:05.842396: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:303] Could not identify NUMA node of platform GPU ID 0, defaulting to 0. Your kernel may not have been built with NUMA support. 2023-11-14 20:58:05.842410: I tensorflow/core/common_runtime/pluggable_device/pluggable_device_factory.cc:269] Created TensorFlow device (/job:localhost/replica:0/task:0/device:GPU:0 with 0 MB memory) -> physical PluggableDevice (device: 0, name: METAL, pci bus id: <undefined>) 2023-11-14 20:58:06.172278: I tensorflow/core/grappler/optimizers/custom_graph_optimizer_registry.cc:114] Plugin optimizer for device_type GPU is enabled. 219/219 [==============================] - 148s 676ms/step - loss: 0.4310 - val_loss: 0.2155 Epoch 2/20 219/219 [==============================] - 143s 651ms/step - loss: 0.1627 - val_loss: 0.1514 Epoch 3/20 219/219 [==============================] - 142s 649ms/step - loss: 0.1462 - val_loss: 0.1488 Epoch 4/20 219/219 [==============================] - 141s 644ms/step - loss: 0.1474 - val_loss: 0.1475 Epoch 5/20 219/219 [==============================] - 142s 649ms/step - loss: 0.1477 - val_loss: 0.1508 Epoch 6/20 219/219 [==============================] - 142s 649ms/step - loss: 0.1006 - val_loss: 0.0617 ... ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62397/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62397/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62396	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62396/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62396/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62396/events	https://github.com/tensorflow/tensorflow/issues/62396	1,993,067,601	I_kwDOArmXAs52y8xR	62,396	ParseError: 172:3 : Message type "object_detection.protos.TrainConfig" has no field named "fine_tune_checkpoint_version"	{ "login": "JT-Studios", "id": 106934268, "node_id": "U_kgDOBl-v_A", "avatar_url": "https://avatars.githubusercontent.com/u/106934268?v=4", "gravatar_id": "", "url": "https://api.github.com/users/JT-Studios", "html_url": "https://github.com/JT-Studios", "followers_url": "https://api.github.com/users/JT-Studios/followers", "following_url": "https://api.github.com/users/JT-Studios/following{/other_user}", "gists_url": "https://api.github.com/users/JT-Studios/gists{/gist_id}", "starred_url": "https://api.github.com/users/JT-Studios/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/JT-Studios/subscriptions", "organizations_url": "https://api.github.com/users/JT-Studios/orgs", "repos_url": "https://api.github.com/users/JT-Studios/repos", "events_url": "https://api.github.com/users/JT-Studios/events{/privacy}", "received_events_url": "https://api.github.com/users/JT-Studios/received_events", "type": "User", "site_admin": false }	[ { "id": 1093464312, "node_id": "MDU6TGFiZWwxMDkzNDY0MzEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:others", "name": "type:others", "color": "159b2e", "default": false, "description": "issues not falling in bug, perfromance, support, build and install or feature" }, { "id": 1097545817, "node_id": "MDU6TGFiZWwxMDk3NTQ1ODE3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:apis", "name": "comp:apis", "color": "0052cc", "default": false, "description": "Highlevel API related issues" } ]	closed	false	{ "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false }	[ { "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false } ]	null	[ "@JT-Studios Could you please make sure that you are using the latest version of OD api and try to remove the \"fine_tune_checkpoint_version\" field from your configuration file ?\r\nIn order to expedite the trouble-shooting process, please provide a code snippet to reproduce the issue reported here. \r\n\r\nThank you!", "I have resolved the issue", "@JT-Studios Thank you for the update!\r\nCould you please move this issue to closed status if it is resolved ?\r\nThank you!", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62396\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62396\">No</a>\n", "@sushreebarsa \r\n\r\nThanks for the help!", "Did a few things and this error is appearing again. Same error. I had to reinstalling my virtual environment due to other errors. I tried to update OB api but now my protobuf is broken.", "Ok I solved this. same solution as last time I'm such and idiot. Thanks for the advice again", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62396\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62396\">No</a>\n", "> I have resolved the issue\r\n\r\ncan you please share your solution , since i am facing the same issue\r\n", "@shravani1115 Yeah Could you please share the code that you are using as well as a shortned version of the error", "> @shravani1115 Yeah Could you please share the code that you are using as well as a shortned version of the error\r\n\r\nYes Sure!\r\n\r\nThis is the error I am facing\r\n\r\nParseError Traceback (most recent call last)\r\nCell In[23], line 1\r\n----> 1 config = config_util.get_configs_from_pipeline_file(CONFIG_PATH)\r\n\r\nFile ~\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\object_detection\\utils\\config_util.py:98, in get_configs_from_pipeline_file(pipeline_config_path, config_override)\r\n 96 with tf.io.gfile.GFile(pipeline_config_path, \"r\") as f:\r\n 97 proto_str = f.read()\r\n---> 98 text_format.Merge(proto_str, pipeline_config)\r\n 99 if config_override:\r\n 100 text_format.Merge(config_override, pipeline_config)\r\n\r\nFile ~\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\google\\protobuf\\text_format.py:719, in Merge(text, message, allow_unknown_extension, allow_field_number, descriptor_pool, allow_unknown_field)\r\n 690 def Merge(text,\r\n 691 message,\r\n 692 allow_unknown_extension=False,\r\n 693 allow_field_number=False,\r\n 694 descriptor_pool=None,\r\n 695 allow_unknown_field=False):\r\n 696 \"\"\"Parses a text representation of a protocol message into a message.\r\n 697 \r\n 698 Like Parse(), but allows repeated values for a non-repeated field, and uses\r\n (...)\r\n 717 ParseError: On text parsing problems.\r\n 718 \"\"\"\r\n--> 719 return MergeLines(\r\n 720 text.split(b'\\n' if isinstance(text, bytes) else u'\\n'),\r\n 721 message,\r\n 722 allow_unknown_extension,\r\n 723 allow_field_number,\r\n 724 descriptor_pool=descriptor_pool,\r\n 725 allow_unknown_field=allow_unknown_field)\r\n\r\nFile ~\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\google\\protobuf\\text_format.py:793, in MergeLines(lines, message, allow_unknown_extension, allow_field_number, descriptor_pool, allow_unknown_field)\r\n 768 \"\"\"Parses a text representation of a protocol message into a message.\r\n 769 \r\n 770 See Merge() for more details.\r\n (...)\r\n 787 ParseError: On text parsing problems.\r\n 788 \"\"\"\r\n 789 parser = _Parser(allow_unknown_extension,\r\n 790 allow_field_number,\r\n 791 descriptor_pool=descriptor_pool,\r\n 792 allow_unknown_field=allow_unknown_field)\r\n--> 793 return parser.MergeLines(lines, message)\r\n\r\nFile ~\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\google\\protobuf\\text_format.py:818, in _Parser.MergeLines(self, lines, message)\r\n 816 \"\"\"Merges a text representation of a protocol message into a message.\"\"\"\r\n 817 self._allow_multiple_scalars = True\r\n--> 818 self._ParseOrMerge(lines, message)\r\n 819 return message\r\n\r\nFile ~\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\google\\protobuf\\text_format.py:837, in _Parser._ParseOrMerge(self, lines, message)\r\n 835 tokenizer = Tokenizer(str_lines)\r\n 836 while not tokenizer.AtEnd():\r\n--> 837 self._MergeField(tokenizer, message)\r\n\r\nFile ~\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\google\\protobuf\\text_format.py:967, in _Parser._MergeField(self, tokenizer, message)\r\n 964 tokenizer.Consume(',')\r\n 966 else:\r\n--> 967 merger(tokenizer, message, field)\r\n 969 else: # Proto field is unknown.\r\n 970 assert (self.allow_unknown_extension or self.allow_unknown_field)\r\n\r\nFile ~\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\google\\protobuf\\text_format.py:1042, in _Parser._MergeMessageField(self, tokenizer, message, field)\r\n 1040 if tokenizer.AtEnd():\r\n 1041 raise tokenizer.ParseErrorPreviousToken('Expected \"%s\".' % (end_token,))\r\n-> 1042 self._MergeField(tokenizer, sub_message)\r\n 1044 if is_map_entry:\r\n 1045 value_cpptype = field.message_type.fields_by_name['value'].cpp_type\r\n\r\nFile ~\\AppData\\Local\\Programs\\Python\\Python310\\lib\\site-packages\\google\\protobuf\\text_format.py:932, in _Parser._MergeField(self, tokenizer, message)\r\n 929 field = None\r\n 931 if not field and not self.allow_unknown_field:\r\n--> 932 raise tokenizer.ParseErrorPreviousToken(\r\n 933 'Message type \"%s\" has no field named \"%s\".' %\r\n 934 (message_descriptor.full_name, name))\r\n 936 if field:\r\n 937 if not self._allow_multiple_scalars and field.containing_oneof:\r\n 938 # Check if there's a different field set in this oneof.\r\n 939 # Note that we ignore the case if the same field was set before, and we\r\n 940 # apply _allow_multiple_scalars to non-scalar fields as well.\r\n\r\nParseError: 172:3 : Message type \"object_detection.protos.TrainConfig\" has no field named \"fine_tune_checkpoint_version\".\r\n\r\n\r\nthis is the code \r\n\r\nimport tensorflow as tf\r\nfrom object_detection.utils import config_util\r\nfrom object_detection.protos import pipeline_pb2\r\nfrom google.protobuf import text_format\r\n\r\nCONFIG_PATH = MODEL_PATH+'/'+CUSTOM_MODEL_NAME+'/pipeline.config'\r\n\r\nconfig = config_util.get_configs_from_pipeline_file(CONFIG_PATH)\r\n\r\nplease help me out\r\n\r\n\r\n\r\n", "@shravani1115 Okay, find your pipeline.config and deleate the line that says fine_tune_checkpoint_version v2 or something along those lines. If this dosn't fix the issue make sure you have the right pipeline.config open as something there can be muytiple. By the where did this code come from?", "Thank you error solved!! I am referring the youtube video \r\n", "> Thank you error solved!! I am referring the youtube video\r\n\r\ncan you tell me how to solve this problem please\r\n" ]	2023-11-14T16:09:22	2024-05-20T12:30:56	2023-11-30T01:49:31	NONE	null	null	null	### Issue type Others ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version v2.15.0-rc0-8-g2a4ec940bac 2.15.0-rc1 ### Custom code Yes ### OS platform and distribution MacOS 14 ### Mobile device _No response_ ### Python version Python 3.9.12 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I tyr to run this script on Jupyter notebook to set my training model: config = config_util.get_configs_from_pipeline_file(files['PIPELINE_CONFIG']) ### Standalone code to reproduce the issue ```shell I am using the popular tensor flow object detection course in 5 hours when I try to run config = config_util.get_configs_from_pipeline_file(files['PIPELINE_CONFIG']) ``` ### Relevant log output ```shell --------------------------------------------------------------------------- ParseError Traceback (most recent call last) Cell In[11], line 1 ----> 1 config = config_util.get_configs_from_pipeline_file(files['PIPELINE_CONFIG']) File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/object_detection/utils/config_util.py:98, in get_configs_from_pipeline_file(pipeline_config_path, config_override) 96 with tf.io.gfile.GFile(pipeline_config_path, "r") as f: 97 proto_str = f.read() ---> 98 text_format.Merge(proto_str, pipeline_config) 99 if config_override: 100 text_format.Merge(config_override, pipeline_config) File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/google/protobuf/text_format.py:719, in Merge(text, message, allow_unknown_extension, allow_field_number, descriptor_pool, allow_unknown_field) 690 def Merge(text, 691 message, 692 allow_unknown_extension=False, 693 allow_field_number=False, 694 descriptor_pool=None, 695 allow_unknown_field=False): 696 """Parses a text representation of a protocol message into a message. 697 698 Like Parse(), but allows repeated values for a non-repeated field, and uses (...) 717 ParseError: On text parsing problems. 718 """ --> 719 return MergeLines( 720 text.split(b'\n' if isinstance(text, bytes) else u'\n'), 721 message, 722 allow_unknown_extension, 723 allow_field_number, 724 descriptor_pool=descriptor_pool, 725 allow_unknown_field=allow_unknown_field) File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/google/protobuf/text_format.py:793, in MergeLines(lines, message, allow_unknown_extension, allow_field_number, descriptor_pool, allow_unknown_field) 768 """Parses a text representation of a protocol message into a message. 769 770 See Merge() for more details. (...) 787 ParseError: On text parsing problems. 788 """ 789 parser = _Parser(allow_unknown_extension, 790 allow_field_number, 791 descriptor_pool=descriptor_pool, 792 allow_unknown_field=allow_unknown_field) --> 793 return parser.MergeLines(lines, message) File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/google/protobuf/text_format.py:818, in _Parser.MergeLines(self, lines, message) 816 """Merges a text representation of a protocol message into a message.""" 817 self._allow_multiple_scalars = True --> 818 self._ParseOrMerge(lines, message) 819 return message File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/google/protobuf/text_format.py:837, in _Parser._ParseOrMerge(self, lines, message) 835 tokenizer = Tokenizer(str_lines) 836 while not tokenizer.AtEnd(): --> 837 self._MergeField(tokenizer, message) File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/google/protobuf/text_format.py:967, in _Parser._MergeField(self, tokenizer, message) 964 tokenizer.Consume(',') 966 else: --> 967 merger(tokenizer, message, field) 969 else: # Proto field is unknown. 970 assert (self.allow_unknown_extension or self.allow_unknown_field) File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/google/protobuf/text_format.py:1042, in _Parser._MergeMessageField(self, tokenizer, message, field) 1040 if tokenizer.AtEnd(): 1041 raise tokenizer.ParseErrorPreviousToken('Expected "%s".' % (end_token,)) -> 1042 self._MergeField(tokenizer, sub_message) 1044 if is_map_entry: 1045 value_cpptype = field.message_type.fields_by_name['value'].cpp_type File ~/Desktop/Jupyter/TFODCourse/tfod/lib/python3.11/site-packages/google/protobuf/text_format.py:932, in _Parser._MergeField(self, tokenizer, message) 929 field = None 931 if not field and not self.allow_unknown_field: --> 932 raise tokenizer.ParseErrorPreviousToken( 933 'Message type "%s" has no field named "%s".' % 934 (message_descriptor.full_name, name)) 936 if field: 937 if not self._allow_multiple_scalars and field.containing_oneof: 938 # Check if there's a different field set in this oneof. 939 # Note that we ignore the case if the same field was set before, and we 940 # apply _allow_multiple_scalars to non-scalar fields as well. ParseError: 172:3 : Message type "object_detection.protos.TrainConfig" has no field named "fine_tune_checkpoint_version". ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62396/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62396/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62395	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62395/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62395/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62395/events	https://github.com/tensorflow/tensorflow/issues/62395	1,992,936,339	I_kwDOArmXAs52ycuT	62,395	No session factory registered for the given session options: {target: "" config: } Registered factories are {}	{ "login": "lidesheng0477", "id": 2074772, "node_id": "MDQ6VXNlcjIwNzQ3NzI=", "avatar_url": "https://avatars.githubusercontent.com/u/2074772?v=4", "gravatar_id": "", "url": "https://api.github.com/users/lidesheng0477", "html_url": "https://github.com/lidesheng0477", "followers_url": "https://api.github.com/users/lidesheng0477/followers", "following_url": "https://api.github.com/users/lidesheng0477/following{/other_user}", "gists_url": "https://api.github.com/users/lidesheng0477/gists{/gist_id}", "starred_url": "https://api.github.com/users/lidesheng0477/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/lidesheng0477/subscriptions", "organizations_url": "https://api.github.com/users/lidesheng0477/orgs", "repos_url": "https://api.github.com/users/lidesheng0477/repos", "events_url": "https://api.github.com/users/lidesheng0477/events{/privacy}", "received_events_url": "https://api.github.com/users/lidesheng0477/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 3797168204, "node_id": "LA_kwDOArmXAs7iVDBM", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.8", "name": "TF 2.8", "color": "5DC9D0", "default": false, "description": "" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[]	2023-11-14T15:06:16	2023-12-06T00:21:34	null	NONE	null	null	null	### Issue type Build/Install ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf2.8.3 ### Custom code Yes ### OS platform and distribution Linux Centos7.5 ### Mobile device _No response_ ### Python version 3.8 ### Bazel version 4.2.4 ### GCC/compiler version 8.3.1 ### CUDA/cuDNN version None ### GPU model and memory None ### Current behavior? 1. Try to create a new TensorFlow session: ``` Session* session; Status status = NewSession(SessionOptions(), &session); ``` 2. Compile succeeds: ``` g++ -o "build64_release/thirdparty/tensorflow-2.8.3/tf_test.objs/tf_test.cc.o" -c -std=gnu++14 -msse4.2 -faligned-new -Wno-invalid-offsetof -Werror=conversion-null -Werror=reorder -Werror=non-virtual-dtor -Werror=overloaded-virtual -fPIC -m64 -mcx16 -pipe -g -DNDEBUG -D_FILE_OFFSET_BITS=64 -D__STDC_CONSTANT_MACROS -D__STDC_FORMAT_MACROS -D__STDC_LIMIT_MACROS -g -Wall -Wextra -Wno-unused-but-set-variable -Wno-unused-parameter -Wno-unused-local-typedefs -Wno-missing-field-initializers -Wno-unused-function -Wendif-labels -Wformat=2 -Wframe-larger-than=69632 -Wmissing-include-dirs -Wpointer-arith -Wwrite-strings -Werror=char-subscripts -Werror=comments -Werror=empty-body -Werror=endif-labels -Werror=format -Werror=missing-include-dirs -Werror=overflow -Werror=parentheses -Werror=return-type -Werror=sequence-point -Werror=sign-compare -Werror=switch -Werror=type-limits -Werror=uninitialized -Werror=unused-label -Werror=unused-result -Werror=unused-value -Werror=unused-variable -Werror=write-strings -Werror=vla -O3 -fno-omit-frame-pointer -Wno-error=overloaded-virtual -lm -Wl,--allow-multiple-definition -Wl,--whole-archive -Wl,--no-as-needed -Ithirdparty/protobuf/src -Ithirdparty -Ibuild64_release -I. -Ithirdparty/tensorflow-2.8.3/include "thirdparty/tensorflow-2.8.3/tf_test.cc" g++ -o "build64_release/thirdparty/tensorflow-2.8.3/tf_test" -m64 -static-libgcc -static-libstdc++ -Wl,--rpath-link=build64_release/thirdparty/tensorflow-2.8.3 build64_release/version.os "build64_release/thirdparty/tensorflow-2.8.3/tf_test.objs/tf_test.cc.o" "build64_release/thirdparty/tensorflow-2.8.3/libtensorflow_framework.so" ``` 3. But running the code gives: ``` 2023-11-14 15:00:31.456030: E tensorflow/core/common_runtime/session.cc:81] Failed to get session factory: NOT_FOUND: No session factory registered for the given session options: {target: "" config: } Registered factories are {}. ``` 4. I have tried as [#3308](https://github.sheincorp.cn/tensorflow/tensorflow/issues/3308) and [#36433](https://github.sheincorp.cn/tensorflow/tensorflow/issues/36433), but it didn't work. ### Standalone code to reproduce the issue ```shell #include "thirdparty/tensorflow-2.8.3/include/tensorflow/core/public/session.h" #include "thirdparty/tensorflow-2.8.3/include/tensorflow/core/platform/env.h" #include "thirdparty/tensorflow-2.8.3/include/tensorflow/cc/client/client_session.h" using namespace tensorflow; /** * @brief deep model for click through rate prediction * @details [long description] * * @param argv[1] graph protobuf * * @return [description] / int main(int argc, char argv[]) { // Initialize a tensorflow session Session* session; Status status = NewSession(SessionOptions(), &session); if (!status.ok()) { std::cerr << status.ToString() << std::endl; return 1; } else { std::cout << "Session created successfully" << std::endl; } // Load the protobuf graph GraphDef graph_def; std::string graph_path = "/develop/testdata/serving_demo/ps/ps1.11.64"; status = ReadBinaryProto(Env::Default(), graph_path, &graph_def); if (!status.ok()) { std::cerr << status.ToString() << std::endl; return 1; } else { std::cout << "Load graph protobuf successfully" << std::endl; } // Add the graph to the session status = session->Create(graph_def); if (!status.ok()) { std::cerr << status.ToString() << std::endl; return 1; } else { std::cout << "Add graph to session successfully" << std::endl; } // Setup inputs and outputs: // Our graph doesn't require any inputs, since it specifies default values, // but we'll change an input to demonstrate. Tensor a(DT_FLOAT, TensorShape()); a.scalar<float>()() = 3.0; Tensor b(DT_FLOAT, TensorShape()); b.scalar<float>()() = 2.0; std::vector<std::pair<string, tensorflow::Tensor>> inputs = { { "a", a }, { "b", b }, }; // The session will initialize the outputs std::vector<tensorflow::Tensor> outputs; // Run the session, evaluating our "c" operation from the graph status = session->Run(inputs, {"c"}, {}, &outputs); if (!status.ok()) { std::cerr << status.ToString() << std::endl; return 1; } else { std::cout << "Run session successfully" << std::endl; } // Grab the first output (we only evaluated one graph node: "c") // and convert the node to a scalar representation. auto output_c = outputs[0].scalar<float>(); // (There are similar methods for vectors and matrices here: // https://github.com/tensorflow/tensorflow/blob/master/tensorflow/core/public/tensor.h) // Print the results std::cout << outputs[0].DebugString() << std::endl; // Tensor<type: float shape: [] values: 30> std::cout << "output value: " << output_c() << std::endl; // 30 // Free any resources used by the session session->Close(); return 0; } ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62395/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62395/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62394	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62394/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62394/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62394/events	https://github.com/tensorflow/tensorflow/pull/62394	1,992,909,663	PR_kwDOArmXAs5fa29v	62,394	[Ignore] Testing CI	{ "login": "MichaelHudgins", "id": 30155094, "node_id": "MDQ6VXNlcjMwMTU1MDk0", "avatar_url": "https://avatars.githubusercontent.com/u/30155094?v=4", "gravatar_id": "", "url": "https://api.github.com/users/MichaelHudgins", "html_url": "https://github.com/MichaelHudgins", "followers_url": "https://api.github.com/users/MichaelHudgins/followers", "following_url": "https://api.github.com/users/MichaelHudgins/following{/other_user}", "gists_url": "https://api.github.com/users/MichaelHudgins/gists{/gist_id}", "starred_url": "https://api.github.com/users/MichaelHudgins/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/MichaelHudgins/subscriptions", "organizations_url": "https://api.github.com/users/MichaelHudgins/orgs", "repos_url": "https://api.github.com/users/MichaelHudgins/repos", "events_url": "https://api.github.com/users/MichaelHudgins/events{/privacy}", "received_events_url": "https://api.github.com/users/MichaelHudgins/received_events", "type": "User", "site_admin": false }	[ { "id": 1169364259, "node_id": "MDU6TGFiZWwxMTY5MzY0MjU5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:XS", "name": "size:XS", "color": "adafea", "default": false, "description": "CL Change Size: Extra Small" } ]	closed	false	null	[]	null	[]	2023-11-14T14:53:28	2023-11-14T16:49:29	2023-11-14T16:49:29	COLLABORATOR	null	true	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62394", "html_url": "https://github.com/tensorflow/tensorflow/pull/62394", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62394.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62394.patch", "merged_at": null }	Ignore - testing CI	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62394/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62394/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62393	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62393/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62393/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62393/events	https://github.com/tensorflow/tensorflow/issues/62393	1,992,452,411	I_kwDOArmXAs52wmk7	62,393	overflow happens in tf.image.adjust_gamma when gamma is very small	{ "login": "drewshark", "id": 128925028, "node_id": "U_kgDOB689ZA", "avatar_url": "https://avatars.githubusercontent.com/u/128925028?v=4", "gravatar_id": "", "url": "https://api.github.com/users/drewshark", "html_url": "https://github.com/drewshark", "followers_url": "https://api.github.com/users/drewshark/followers", "following_url": "https://api.github.com/users/drewshark/following{/other_user}", "gists_url": "https://api.github.com/users/drewshark/gists{/gist_id}", "starred_url": "https://api.github.com/users/drewshark/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/drewshark/subscriptions", "organizations_url": "https://api.github.com/users/drewshark/orgs", "repos_url": "https://api.github.com/users/drewshark/repos", "events_url": "https://api.github.com/users/drewshark/events{/privacy}", "received_events_url": "https://api.github.com/users/drewshark/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false }	[ { "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false } ]	null	[ "Hey @drewshark! can I work on this issue?", "@drewshark I was able to replicate this issue. As a workaround could you try to use a smaller value of gamma or the algorithm which doesn't include floating point numbers. Please let us know?\r\nThank you!", "Hi @sushreebarsa , sure I will use floating tensor instead.", "@drewshark Thank you for your response here!\r\nCould you please let us know if we can move this issue to closed status if the workaround worked for you ?\r\nThank you!", "Sure, I am closing this issue.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62393\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62393\">No</a>\n" ]	2023-11-14T10:31:13	2023-11-24T08:51:30	2023-11-24T08:51:27	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.14.0 ### Custom code Yes ### OS platform and distribution _No response_ ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? When `gamma` is a very small value (e.g., 0.00001 or 0) and input tensor is a positive integer tensor, tf.image.adjust_gamma might output largely negative value. Specifically, if gamma=0 and gain=2, the expected output should be 2 according to the formula: `Out = gain * In**gamma`, but the current result is -9223372036854775808. It seems that the overflow issue happens when executing this code: https://github.com/tensorflow/tensorflow/blob/9fa48d09da86703782f68c523327ad57c2cca8ed/tensorflow/python/ops/image_ops_impl.py#L2372 ### Standalone code to reproduce the issue ```shell import tensorflow as tf image = tf.constant(9, dtype='int64') gamma = 0 gain = 2 out = tf.image.adjust_gamma(image,gamma,gain) print(out) ``` ### Relevant log output ```shell tf.Tensor(-9223372036854775808, shape=(), dtype=int64) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62393/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62393/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62392	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62392/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62392/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62392/events	https://github.com/tensorflow/tensorflow/issues/62392	1,992,417,883	I_kwDOArmXAs52weJb	62,392	TfLite 2.14 fails to build with Visual Studio with AVX2 compiler flag	{ "login": "misterBart", "id": 54583863, "node_id": "MDQ6VXNlcjU0NTgzODYz", "avatar_url": "https://avatars.githubusercontent.com/u/54583863?v=4", "gravatar_id": "", "url": "https://api.github.com/users/misterBart", "html_url": "https://github.com/misterBart", "followers_url": "https://api.github.com/users/misterBart/followers", "following_url": "https://api.github.com/users/misterBart/following{/other_user}", "gists_url": "https://api.github.com/users/misterBart/gists{/gist_id}", "starred_url": "https://api.github.com/users/misterBart/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/misterBart/subscriptions", "organizations_url": "https://api.github.com/users/misterBart/orgs", "repos_url": "https://api.github.com/users/misterBart/repos", "events_url": "https://api.github.com/users/misterBart/events{/privacy}", "received_events_url": "https://api.github.com/users/misterBart/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 1188421838, "node_id": "MDU6TGFiZWwxMTg4NDIxODM4", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:windows", "name": "subtype:windows", "color": "b619ea", "default": false, "description": "Windows Build/Installation Issues" }, { "id": 5922361893, "node_id": "LA_kwDOArmXAs8AAAABYQASJQ", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF2.14", "name": "TF2.14", "color": "b60205", "default": false, "description": "For issues related to Tensorflow 2.14.x" } ]	closed	false	{ "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false }	[ { "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @misterBart,\r\n\r\nI didn't get the error messages with nightly:\r\n\r\n```\r\ngit clone --single-branch --branch nightly https://github.com/tensorflow/tensorflow tensorflow_src\r\nmkdir tflite_x64_release\r\ncd tflite_x64_release\r\ncmake -G \"Visual Studio 17 2022\" -A x64 -DCMAKE_C_FLAGS_INIT=\"/arch:AVX2\" -DCMAKE_CXX_FLAGS_INIT=\"/arch:AVX2\" ..\\tensorflow_src\\tensorflow\\lite\r\ncmake --build . -j 8 --config release\r\n```\r\n\r\nseems like abseil is updated after 2.14 release, can you use nightly to progress?", "You're right, I do not receive the reported error with the nightly branch. \r\n\r\nI am receiving a different error now though, haha:\r\n`C:\\Users\\bartp\\source\\TfNightly-Attempt2\\tensorflow_src\\tensorflow/lite/kernels/internal/optimized/fully_connected_4bit.h(21,10): fatal error C1083: Cannot open include file: 'sys/mman.h': No such file or directory [C:\\Users\\bartp\\source\\TfNightly-Attempt2\\tflite_x64_release\\tensorflow-lite.vcxproj]`\r\nBut that is different issue, so I will close this issue. \r\n\r\nThanks for helping out.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62392\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62392\">No</a>\n" ]	2023-11-14T10:11:08	2023-11-22T17:31:49	2023-11-22T17:31:45	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version tf 2.14 ### Custom code No ### OS platform and distribution Windows 10 Enterprise ### Mobile device _No response_ ### Python version _No response_ ### Bazel version _No response_ ### GCC/compiler version Visual Studio 2019 and 2022 ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Building TfLite 2.14 yields the following build error with Visual Studio 2022 (and 2019) when the AVX2 compiler flag is enabled: ``` C:\Users\bartp\source\TfLite2.14\tflite_x64_release\abseil-cpp\absl\crc\internal\crc_memcpy_x86_64.cc(391,11): error C7555: use of designated initializers requires at least '/std:c++20' [C:\Users\bartp\source\TfLite2.14\tflite_x64_release\_deps\abseil-cpp-build\absl\crc\absl_crc32c.vcxproj] C:\Users\bartp\source\TfLite2.14\tflite_x64_release\abseil-cpp\absl\crc\internal\crc_memcpy_x86_64.cc(403,11): error C7555: use of designated initializers requires at least '/std:c++20' [C:\Users\bartp\source\TfLite2.14\tflite_x64_release\_deps\abseil-cpp-build\absl\crc\absl_crc32c.vcxproj] C:\Users\bartp\source\TfLite2.14\tflite_x64_release\abseil-cpp\absl\crc\internal\crc_memcpy_x86_64.cc(409,11): error C7555: use of designated initializers requires at least '/std:c++20' [C:\Users\bartp\source\TfLite2.14\tflite_x64_release\_deps\abseil-cpp-build\absl\crc\absl_crc32c.vcxproj] C:\Users\bartp\source\TfLite2.14\tflite_x64_release\abseil-cpp\absl\crc\internal\crc_memcpy_x86_64.cc(413,15): error C7555: use of designated initializers requires at least '/std:c++20' [C:\Users\bartp\source\TfLite2.14\tflite_x64_release\_deps\abseil-cpp-build\absl\crc\absl_crc32c.vcxproj] ``` This Abseil issue has been solved according to https://github.com/abseil/abseil-cpp/issues/1504 Seemingly TfLite 2.14 uses an unpatched version of Abseil. ### Standalone code to reproduce the issue ```shell In Command Prompt execute: git clone --single-branch --branch r2.14 https://github.com/tensorflow/tensorflow tensorflow_src mkdir tflite_x64_release cd tflite_x64_release cmake -G "Visual Studio 17 2022" -A x64 -DCMAKE_C_FLAGS_INIT="/arch:AVX2" -DCMAKE_CXX_FLAGS_INIT="/arch:AVX2" ..\tensorflow_src\tensorflow\lite cmake --build . -j 8 --config release ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62392/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62392/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62391	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62391/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62391/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62391/events	https://github.com/tensorflow/tensorflow/issues/62391	1,992,181,358	I_kwDOArmXAs52vkZu	62,391	jit-compiled `tfnp.take_along_axis` shape bug	{ "login": "jackd", "id": 659115, "node_id": "MDQ6VXNlcjY1OTExNQ==", "avatar_url": "https://avatars.githubusercontent.com/u/659115?v=4", "gravatar_id": "", "url": "https://api.github.com/users/jackd", "html_url": "https://github.com/jackd", "followers_url": "https://api.github.com/users/jackd/followers", "following_url": "https://api.github.com/users/jackd/following{/other_user}", "gists_url": "https://api.github.com/users/jackd/gists{/gist_id}", "starred_url": "https://api.github.com/users/jackd/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/jackd/subscriptions", "organizations_url": "https://api.github.com/users/jackd/orgs", "repos_url": "https://api.github.com/users/jackd/repos", "events_url": "https://api.github.com/users/jackd/events{/privacy}", "received_events_url": "https://api.github.com/users/jackd/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 1097547147, "node_id": "MDU6TGFiZWwxMDk3NTQ3MTQ3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:ops", "name": "comp:ops", "color": "0052cc", "default": false, "description": "OPs related issues" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi, @jackd!\r\nI was able to replicate the issue with jit-compiled=True but the issue is not reproducible with jit-compiled=False both in 2.15 and tf-nightly. \r\nPlease have a look at this [gist](https://colab.research.google.com/gist/sushreebarsa/19bb1d29524258b41fce7a7784dcc4ba/62391.ipynb#scrollTo=X-sMyV14FWPP) for reference. \r\nAs a workaround you can use tf.function(jit_compile=False) or by not using tfnp.take_along_axis altogether to avoid such errors. Could you please confirm the same?\r\nThank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "@sushreebarsa confirmed", "@jackd Thanks for the confirmation. If the workaround works for you then can we move this issue to closed status?\r\nThank you!", "The existence of a workaround isn't the same as a bug fix...", "@sachinprasadhs Could you please have a look at this issue?\r\nThank you!" ]	2023-11-14T07:54:57	2023-12-06T21:39:21	null	CONTRIBUTOR	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source binary ### TensorFlow version tf_nightly-2.16.0.dev20231113-cp310-cp310-manylinux_2_17_x86_64.manylinux2014_x86_64 ### Custom code Yes ### OS platform and distribution colab ### Mobile device _No response_ ### Python version 3.10 ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? [colab demonstrating the issue](https://colab.research.google.com/drive/1nk6NXUoo7qE7g6NHlFHBgMWkJW0LuTMx?usp=sharing) `tf.experimental.numpy.take_along_axis` returns tensor with incorrect shape when used with `tf.function(jit_compile=True)`. ### Standalone code to reproduce the issue ```shell import tensorflow as tf x = tf.random.normal((5, 3, 2)) indices = tf.constant([[[-1]]], dtype="int32") def f(x, i): return tf.squeeze(tf.experimental.numpy.take_along_axis(x, i, axis=-2), axis=-2) z = f(x, indices) print(z.shape) # (5, 2) z1 = tf.function(f, jit_compile=True)(x, indices) # errors print(z1.shape) ``` ### Relevant log output ```shell --------------------------------------------------------------------------- InvalidArgumentError Traceback (most recent call last) <ipython-input-4-97d5eab12d23> in <cell line: 1>() ----> 1 z1 = tf.function(f, jit_compile=True)(x, indices) 2 print(z1.shape) 1 frames /usr/local/lib/python3.10/dist-packages/tensorflow/python/eager/execute.py in quick_execute(op_name, num_outputs, inputs, attrs, ctx, name) 51 try: 52 ctx.ensure_initialized() ---> 53 tensors = pywrap_tfe.TFE_Py_Execute(ctx._handle, device_name, op_name, 54 inputs, attrs, num_outputs) 55 except core._NotOkStatusException as e: InvalidArgumentError: Tried to explicitly squeeze dimension 1 but dimension was not 1: 2 Stack trace for op definition: File "/usr/lib/python3.10/runpy.py", line 196, in _run_module_as_main File "/usr/lib/python3.10/runpy.py", line 86, in _run_code File "/usr/local/lib/python3.10/dist-packages/colab_kernel_launcher.py", line 37, in <module> File "/usr/local/lib/python3.10/dist-packages/traitlets/config/application.py", line 992, in launch_instance File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelapp.py", line 619, in start File "/usr/local/lib/python3.10/dist-packages/tornado/platform/asyncio.py", line 195, in start File "/usr/lib/python3.10/asyncio/base_events.py", line 603, in run_forever File "/usr/lib/python3.10/asyncio/base_events.py", line 1909, in _run_once File "/usr/lib/python3.10/asyncio/events.py", line 80, in _run File "/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py", line 685, in <lambda> File "/usr/local/lib/python3.10/dist-packages/tornado/ioloop.py", line 738, in _run_callback File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 825, in inner File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 786, in run File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 361, in process_one File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 261, in dispatch_shell File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/kernelbase.py", line 539, in execute_request File "/usr/local/lib/python3.10/dist-packages/tornado/gen.py", line 234, in wrapper File "/usr/local/lib/python3.10/dist-packages/ipykernel/ipkernel.py", line 302, in do_execute File "/usr/local/lib/python3.10/dist-packages/ipykernel/zmqshell.py", line 539, in run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 2975, in run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3030, in _run_cell File "/usr/local/lib/python3.10/dist-packages/IPython/core/async_helpers.py", line 78, in _pseudo_sync_runner File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3257, in run_cell_async File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3473, in run_ast_nodes File "/usr/local/lib/python3.10/dist-packages/IPython/core/interactiveshell.py", line 3553, in run_code File "<ipython-input-4-97d5eab12d23>", line 1, in <cell line: 1> File "<ipython-input-2-4f365be98111>", line 8, in f [[{{node Squeeze}}]] tf2xla conversion failed while converting __inference_f_283[_XlaMustCompile=true,config_proto=3175580994766145631,executor_type=11160318154034397263]. Run with TF_DUMP_GRAPH_PREFIX=/path/to/dump/dir and --vmodule=xla_compiler=2 to obtain a dump of the compiled functions. [Op:__inference_f_283] ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62391/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62391/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62390	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62390/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62390/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62390/events	https://github.com/tensorflow/tensorflow/pull/62390	1,992,057,993	PR_kwDOArmXAs5fX-f2	62,390	TFLite GPU: Remove all 16-component vectors	{ "login": "looi", "id": 1855766, "node_id": "MDQ6VXNlcjE4NTU3NjY=", "avatar_url": "https://avatars.githubusercontent.com/u/1855766?v=4", "gravatar_id": "", "url": "https://api.github.com/users/looi", "html_url": "https://github.com/looi", "followers_url": "https://api.github.com/users/looi/followers", "following_url": "https://api.github.com/users/looi/following{/other_user}", "gists_url": "https://api.github.com/users/looi/gists{/gist_id}", "starred_url": "https://api.github.com/users/looi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/looi/subscriptions", "organizations_url": "https://api.github.com/users/looi/orgs", "repos_url": "https://api.github.com/users/looi/repos", "events_url": "https://api.github.com/users/looi/events{/privacy}", "received_events_url": "https://api.github.com/users/looi/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169365494, "node_id": "MDU6TGFiZWwxMTY5MzY1NDk0", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:M", "name": "size:M", "color": "adafea", "default": false, "description": "CL Change Size: Medium" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "Note: fixed a small build issue", "Can you provide some comprehensive benchmark values for FC & Conv Transpose? These are very important ops, and all changes to them must be done carefully ", "Hi @looi Can you please check @grantjensen's [comments](https://github.com/tensorflow/tensorflow/pull/62390#issuecomment-1819812742) and keep us posted ? Thank you!", "This PR is stale because it has been open for 14 days with no activity. It will be closed if no further activity occurs. Thank you.", "Sorry, i will post an update soon", "Hi @looi Any update on this PR? Please. Thank you!", "I've tested this change on the model [ssd_mobiledet_cpu_coco](https://github.com/tensorflow/models/blob/master/research/object_detection/g3doc/tf1_detection_zoo.md) (without TFLite_Detection_PostProcess). This model contains FC and Conv Transpose ops. It is originally intended for running on CPU, but it runs fine using the tflite GPU plugin. For the purpose of testing, I also overrode `UseBufferForWeights` in [fully_connected.cc](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/delegates/gpu/common/tasks/fully_connected.cc) and [convolution_transposed.cc](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/delegates/gpu/common/tasks/convolution_transposed.cc) to return true. This change doesn't actually affect any device where that function returns false.\r\n\r\nThe performance is the same or slightly better on these devices:\r\n\r\n* Pixel 3 (Adreno 630): 30ms -> 30ms\r\n* Pixel 7a (Mali-G710): 25ms -> 24ms\r\n* Intel UHD Graphics 630: 10ms -> 10ms\r\n\r\nBasically, this change replaces memory accesses with 16-component vectors (float16) with accesses using 4-component vectors. I expect this change will produce a small but positive effect in performance, because [clpeak](https://github.com/krrishnarraj/clpeak) shows that memory bandwidth for float4 is generally same or better than float16 on mobile platforms. This is probably because float16 is not commonly used (nor is it supported by openGL/vulkan/directx/metal), and drivers are not optimized for it. For example, these are clpeak results:\r\n\r\nPixel 3 (Adreno 630)\r\n```\r\nPlatform: QUALCOMM Snapdragon(TM)\r\n Device: QUALCOMM Adreno(TM)\r\n Driver version : OpenCL 2.0 QUALCOMM build: commit #781e7d0 changeid #I46ff5fc46f Date: 12/01/20 Tue Local Branch: QPR1 Remote Branch: Compiler E031.37.10.00 (Android)\r\n Compute units : 2\r\n Clock frequency : 1 MHz\r\n\r\n Global memory bandwidth (GBPS)\r\n float : 22.20\r\n float2 : 23.42\r\n float4 : 24.34\r\n float8 : 22.94\r\n float16 : 21.73\r\n```\r\n\r\nPixel 7a (Mali-G710)\r\n```\r\nPlatform: ARM Platform\r\n Device: Mali-G710 r0p0\r\n Driver version : 3.0 (Android)\r\n Compute units : 7\r\n Clock frequency : 5 MHz\r\n\r\n Global memory bandwidth (GBPS)\r\n float : 36.29\r\n float2 : 34.98\r\n float4 : 34.59\r\n float8 : 33.64\r\n float16 : 19.63\r\n```\r\n\r\nIntel UHD Graphics 630\r\n```\r\nPlatform: Intel(R) OpenCL HD Graphics\r\n Device: Intel(R) UHD Graphics [0x9bc4]\r\n Driver version : 1.0.0 (Linux x64)\r\n Compute units : 24\r\n Clock frequency : 1150 MHz\r\n\r\n Global memory bandwidth (GBPS)\r\n float : 20.61\r\n float2 : 23.38\r\n float4 : 22.70\r\n float8 : 23.89\r\n float16 : 23.55\r\n```\r\n\r\nPlease let me know if this is helpful, or if you desire other information. Thanks!", "Hi @grantjensen Can you please review this PR ? Thank you!", "Yeah, overall I like this change. This is a nice little find :). Speed performance values look good. The one remaining concern I have is the memory performance. I noticed the Pixel 7a values you provided float4 & float16 differ dramatically. Could you see what the memory looks like for benchmarking a model w & w/o your change on a Pixel 7a?", "For the model above, I observed a small improvement on Pixel 7a (25ms -> 24ms). Compared to Pixel 3 and Intel UHD 630, this is the only device where I observed a consistent (but small) improvement, probably because of the float16 performance difference shown by clpeak. Is there another kind of benchmark you are looking for?", "Nah that looks good to me. Last question though, why remove ReplaceAllWords as well? ", "> Nah that looks good to me. Last question though, why remove ReplaceAllWords as well?\r\n\r\nThe function in that file appears to be unused now. The same or similar function called ReplaceAllWords seems to be duplicated in other files as well, like https://github.com/tensorflow/tensorflow/blob/master/tensorflow/lite/delegates/gpu/metal/metal_arguments.cc" ]	2023-11-14T06:11:41	2024-02-07T22:06:06	2024-02-07T22:06:06	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62390", "html_url": "https://github.com/tensorflow/tensorflow/pull/62390", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62390.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62390.patch", "merged_at": "2024-02-07T22:06:06" }	This simplifies the code, removes the text replacement hack for metal, and probably provides the same or better performance on all platforms. It avoids the need for similar hacks in other platforms too, because 16-component vectors are not commonly supported.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62390/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62390/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62389	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62389/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62389/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62389/events	https://github.com/tensorflow/tensorflow/issues/62389	1,992,027,587	I_kwDOArmXAs52u-3D	62,389	tensorflow.org/versions points to latest version for 2.11 onwards	{ "login": "ageron", "id": 76661, "node_id": "MDQ6VXNlcjc2NjYx", "avatar_url": "https://avatars.githubusercontent.com/u/76661?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ageron", "html_url": "https://github.com/ageron", "followers_url": "https://api.github.com/users/ageron/followers", "following_url": "https://api.github.com/users/ageron/following{/other_user}", "gists_url": "https://api.github.com/users/ageron/gists{/gist_id}", "starred_url": "https://api.github.com/users/ageron/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ageron/subscriptions", "organizations_url": "https://api.github.com/users/ageron/orgs", "repos_url": "https://api.github.com/users/ageron/repos", "events_url": "https://api.github.com/users/ageron/events{/privacy}", "received_events_url": "https://api.github.com/users/ageron/received_events", "type": "User", "site_admin": false }	[ { "id": 284443156, "node_id": "MDU6TGFiZWwyODQ0NDMxNTY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:docs-bug", "name": "type:docs-bug", "color": "159b2e", "default": false, "description": "Document issues" }, { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 1093464312, "node_id": "MDU6TGFiZWwxMDkzNDY0MzEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:others", "name": "type:others", "color": "159b2e", "default": false, "description": "issues not falling in bug, perfromance, support, build and install or feature" } ]	open	false	{ "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false }	[ { "login": "tilakrayal", "id": 81610181, "node_id": "MDQ6VXNlcjgxNjEwMTgx", "avatar_url": "https://avatars.githubusercontent.com/u/81610181?v=4", "gravatar_id": "", "url": "https://api.github.com/users/tilakrayal", "html_url": "https://github.com/tilakrayal", "followers_url": "https://api.github.com/users/tilakrayal/followers", "following_url": "https://api.github.com/users/tilakrayal/following{/other_user}", "gists_url": "https://api.github.com/users/tilakrayal/gists{/gist_id}", "starred_url": "https://api.github.com/users/tilakrayal/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/tilakrayal/subscriptions", "organizations_url": "https://api.github.com/users/tilakrayal/orgs", "repos_url": "https://api.github.com/users/tilakrayal/repos", "events_url": "https://api.github.com/users/tilakrayal/events{/privacy}", "received_events_url": "https://api.github.com/users/tilakrayal/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @ageron ,\r\n\r\nI have cross checked and acknowledged the issue. The script seems not public. I will raise an internal PR to fix this. Thanks for reporting.", "@ageron,\r\nThere was an internal issue raised for tracking the same issue and it got resolved now. Could you please try to access any Keras in the Tensorflow.org page, now it is able to re-direct to correct source code. Please check and confirm if it is working in your case.\r\n\r\nhttps://www.tensorflow.org/api_docs/python/tf ------ > click on r2.11 or r2.12------>it is redirecting the right page.\r\n\r\nhttps://www.tensorflow.org/api_docs/python/tf/keras/Model\r\nhttps://github.com/keras-team/keras/blob/v3.3.3/keras/src/models/model.py#L32-L548\r\n\r\nThank you!" ]	2023-11-14T05:40:24	2024-06-12T11:03:09	null	CONTRIBUTOR	null	null	null	### Issue type Documentation Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 2.11+ ### Custom code No ### OS platform and distribution All ### Mobile device All ### Python version All ### Bazel version _No response_ ### GCC/compiler version _No response_ ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? In https://tensorflow.org/versions, the links to versions 2.11, 2.12, 2.13, and 2.14 all point to the latest docs (i.e., https://www.tensorflow.org/api_docs/python/tf) instead of pointing to https://www.tensorflow.org/versions/r2.11/api_docs/python/tf (or to a similar URL for 2.12, and 2.13). If the script that generates this page is available on github, I'm happy to fix it, but I just can't find it. Note that the links to the release notes are okay. ### Standalone code to reproduce the issue ```shell Step 1: go to https://tensorflow.org/versions Step 2: click on r2.11, r2.12, or r2.13. All of these lead to the wrong page. Once 2.15 is released, I'm guessing r2.14 will also be wrong (it will point to the latest docs). ``` ### Relevant log output ```shell N/A ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62389/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62389/timeline	null	null	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62388	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62388/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62388/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62388/events	https://github.com/tensorflow/tensorflow/issues/62388	1,991,937,810	I_kwDOArmXAs52uo8S	62,388	windows build tensorflowlite for android error	{ "login": "leizhu1989", "id": 35773005, "node_id": "MDQ6VXNlcjM1NzczMDA1", "avatar_url": "https://avatars.githubusercontent.com/u/35773005?v=4", "gravatar_id": "", "url": "https://api.github.com/users/leizhu1989", "html_url": "https://github.com/leizhu1989", "followers_url": "https://api.github.com/users/leizhu1989/followers", "following_url": "https://api.github.com/users/leizhu1989/following{/other_user}", "gists_url": "https://api.github.com/users/leizhu1989/gists{/gist_id}", "starred_url": "https://api.github.com/users/leizhu1989/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/leizhu1989/subscriptions", "organizations_url": "https://api.github.com/users/leizhu1989/orgs", "repos_url": "https://api.github.com/users/leizhu1989/repos", "events_url": "https://api.github.com/users/leizhu1989/events{/privacy}", "received_events_url": "https://api.github.com/users/leizhu1989/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473184161, "node_id": "MDU6TGFiZWw0NzMxODQxNjE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:support", "name": "type:support", "color": "159b2e", "default": false, "description": "Support issues" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 5206407904, "node_id": "LA_kwDOArmXAs8AAAABNlN64A", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.12", "name": "TF 2.12", "color": "c5def5", "default": false, "description": "For issues related to Tensorflow 2.12" } ]	closed	false	{ "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false }	[ { "login": "pkgoogle", "id": 132095473, "node_id": "U_kgDOB9-d8Q", "avatar_url": "https://avatars.githubusercontent.com/u/132095473?v=4", "gravatar_id": "", "url": "https://api.github.com/users/pkgoogle", "html_url": "https://github.com/pkgoogle", "followers_url": "https://api.github.com/users/pkgoogle/followers", "following_url": "https://api.github.com/users/pkgoogle/following{/other_user}", "gists_url": "https://api.github.com/users/pkgoogle/gists{/gist_id}", "starred_url": "https://api.github.com/users/pkgoogle/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/pkgoogle/subscriptions", "organizations_url": "https://api.github.com/users/pkgoogle/orgs", "repos_url": "https://api.github.com/users/pkgoogle/repos", "events_url": "https://api.github.com/users/pkgoogle/events{/privacy}", "received_events_url": "https://api.github.com/users/pkgoogle/received_events", "type": "User", "site_admin": false } ]	null	[ "@leizhu1989 Could you please make sure that the JDK is installed and that it is installed in the location that is specified in the @local_jdk//:jdk target. You can check the location of the JDK by running the following command:\r\n```\r\njava -version\r\n```\r\n\r\nThank you!", "@sushreebarsa thank you for your reply ! when I add --java_runtime_version=remotejdk_11 in build command, it has no that error,but also has a error like:\r\n\r\nINFO: Found applicable config definition build:short_logs in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --output_filter=DONT_MATCH_ANYTHING\r\nINFO: Found applicable config definition build:v2 in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1\r\nINFO: Found applicable config definition build:windows in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --copt=/W0 --host_copt=/W0 --copt=/Zc:__cplusplus --host_copt=/Zc:__cplusplus --copt=/D_USE_MATH_DEFINES --host_copt=/D_USE_MATH_DEFINES --features=compiler_param_file --copt=/d2ReducedOptimizeHugeFunctions --host_copt=/d2ReducedOptimizeHugeFunctions --cxxopt=/std:c++17 --host_cxxopt=/std:c++17 --config=monolithic --copt=-DWIN32_LEAN_AND_MEAN --host_copt=-DWIN32_LEAN_AND_MEAN --copt=-DNOGDI --host_copt=-DNOGDI --copt=/Zc:preprocessor --host_copt=/Zc:preprocessor --linkopt=/DEBUG --host_linkopt=/DEBUG --linkopt=/OPT:REF --host_linkopt=/OPT:REF --linkopt=/OPT:ICF --host_linkopt=/OPT:ICF --verbose_failures --features=compiler_param_file --distinct_host_configuration=false\r\nINFO: Found applicable config definition build:monolithic in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --define framework_shared_object=false --define tsl_protobuf_header_only=false --experimental_link_static_libraries_once=false\r\nWARNING: Download from https://mirror.bazel.build/github.com/bazelbuild/rules_cc/archive/081771d4a0e9d7d3aa0eed2ef389fa4700dfb23e.tar.gz failed: class java.io.FileNotFoundException GET returned 404 Not Found\r\nWARNING: Download from https://storage.googleapis.com/mirror.tensorflow.org/github.com/google/XNNPACK/archive/659147817805d17c7be2d60bd7bbca7e780f9c82.zip failed: class java.io.FileNotFoundException GET returned 404 Not Found\r\nINFO: Analyzed target //tensorflow/lite/java:tensorflowlite (0 packages loaded, 0 targets configured).\r\nINFO: Found 1 target...\r\nERROR: D:/project/deeplearning/tensorflow/tensorflow/lite/nnapi/BUILD:31:11: Compiling tensorflow/lite/nnapi/nnapi_implementation.cc failed: (Exit 1): clang failed: error executing command\r\n cd /d C:/users/zhulei/_bazel_zhulei/qotdwtts/execroot/org_tensorflow\r\n SET ANDROID_BUILD_TOOLS_VERSION=34.0.0\r\n SET ANDROID_NDK_API_LEVEL=26\r\n SET ANDROID_NDK_HOME=D:/project/android-ndk-r20b-windows-x86_64\r\n SET ANDROID_SDK_API_LEVEL=34\r\n SET ANDROID_SDK_HOME=D:/Android/Sdk\r\n SET PATH=D:\\project\\install_soft\\msys64\\usr\\bin;D:\\project\\install_soft\\msys64\\bin;C:\\WINDOWS;C:\\WINDOWS\\System32;C:\\WINDOWS\\System32\\WindowsPowerShell\\v1.0;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.4\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.4\\libnvvp;C:\\Program Files (x86)\\VMware\\VMware Workstation\\bin\\;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.1\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.1\\libnvvp;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.3\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.3\\libnvvp;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v10.0\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v10.0\\libnvvp;C:\\windows\\system32;C:\\windows;C:\\windows\\System32\\Wbem;C:\\windows\\System32\\WindowsPowerShell\\v1.0\\;C:\\windows\\System32\\OpenSSH\\;C:\\Program Files (x86)\\NVIDIA Corporation\\PhysX\\Common;C:\\Program Files\\NVIDIA Corporation\\NVIDIA NvDLISR;C:\\Windows\\system32;C:\\Windows;C:\\Windows\\System32\\Wbem;C:\\Windows\\System32\\WindowsPowerShell\\v1.0\\;C:\\Windows\\System32\\OpenSSH\\;C:\\Program Files (x86)\\NetSarang\\Xftp 7\\;C:\\Program Files (x86)\\PuTTY\\;C:\\Users\\zhulei\\.dnx\\bin;C:\\Program Files\\Microsoft DNX\\Dnvm\\;C:\\Program Files\\Microsoft SQL Server\\130\\Tools\\Binn\\;C:\\Program Files\\Git\\cmd;C:\\ProgramData\\Anaconda3;C:\\ProgramData\\Anaconda3\\Library\\mingw-w64\\bin;C:\\ProgramData\\Anaconda3\\Library\\bin;C:\\ProgramData\\Anaconda3\\Scripts;E:\\project\\cmake-3.20.0-rc3-windows-x86_64\\bin;C:\\project\\platform-tools_r33.0.3-windows\\platform-tools;E:\\project\\protobuf-3.4.0-gmock\\install\\protobuf\\bin;C:\\Program Files\\IDM Computer Solutions\\UltraEdit;D:\\project\\TensorRT-8.4.2.4.Windows10.x86_64.cuda-11.6.cudnn8.4\\lib;C:\\Program Files\\NVIDIA Corporation\\Nsight Compute 2021.2.2\\;D:\\project\\Elasticsearch\\elasticsearch-8.1.2\\jdk;D:\\project\\Elasticsearch\\node-v17.9.1-win-x64;C:\\WINDOWS\\system32;C:\\WINDOWS;C:\\WINDOWS\\System32\\Wbem;C:\\WINDOWS\\System32\\WindowsPowerShell\\v1.0\\;C:\\WINDOWS\\System32\\OpenSSH\\;C:\\Program Files (x86)\\NetSarang\\Xshell 7\\;D:\\project\\cudnn-windows-x86_64-8.4.0.27_cuda11.6-archive\\bin;D:\\project\\install_soft;D:\\project\\install_soft\\msys64\\usr\\bin;C:\\Users\\zhulei\\AppData\\Local\\Microsoft\\WindowsApps;;C:\\Users\\zhulei\\AppData\\Local\\Programs\\Microsoft VS Code\\bin\r\n SET PWD=/proc/self/cwd\r\n SET PYTHON_BIN_PATH=C:/ProgramData/Anaconda3/python.exe\r\n SET PYTHON_LIB_PATH=C:/ProgramData/Anaconda3/lib/site-packages\r\n SET RUNFILES_MANIFEST_ONLY=1\r\n SET TF2_BEHAVIOR=1\r\n external\\androidndk\\ndk\\toolchains\\llvm\\prebuilt\\windows-x86_64\\bin\\clang -D__ANDROID_API__=26 -isystemexternal/androidndk/ndk/sysroot/usr/include/arm-linux-androideabi -target armv7-none-linux-androideabi -march=armv7-a -mfloat-abi=softfp -mfpu=vfpv3-d16 -gcc-toolchain external/androidndk/ndk/toolchains/arm-linux-androideabi-4.9/prebuilt/windows-x86_64 -fpic -no-canonical-prefixes -Wno-invalid-command-line-argument -Wno-unused-command-line-argument -funwind-tables -fstack-protector-strong -fno-addrsig -Werror=return-type -Werror=int-to-pointer-cast -Werror=pointer-to-int-cast -Werror=implicit-function-declaration -mthumb -Os -g -DNDEBUG -MD -MF bazel-out/armeabi-v7a-opt/bin/tensorflow/lite/nnapi/_objs/nnapi_implementation/nnapi_implementation.pic.d -frandom-seed=bazel-out/armeabi-v7a-opt/bin/tensorflow/lite/nnapi/_objs/nnapi_implementation/nnapi_implementation.pic.o -fPIC -iquote . -iquote bazel-out/armeabi-v7a-opt/bin /W0 /Zc:__cplusplus /D_USE_MATH_DEFINES /d2ReducedOptimizeHugeFunctions -DWIN32_LEAN_AND_MEAN -DNOGDI /Zc:preprocessor /d2ReducedOptimizeHugeFunctions /std:c++17 '--std=c++11' --sysroot=external/androidndk/ndk/platforms/android-26/arch-arm -isystem external/androidndk/ndk/sources/cxx-stl/llvm-libc++/include -isystem external/androidndk/ndk/sources/cxx-stl/llvm-libc++abi/include -isystem external/androidndk/ndk/sources/android/support/include -isystemexternal/androidndk/ndk/sysroot/usr/include -c tensorflow/lite/nnapi/nnapi_implementation.cc -o bazel-out/armeabi-v7a-opt/bin/tensorflow/lite/nnapi/_objs/nnapi_implementation/nnapi_implementation.pic.o\r\n# Configuration: 3dcec145ccee737eeff39455689d10f429749c273e882c1f289971a54a4328c4\r\n# Execution platform: @local_execution_config_platform//:platform\r\nclang: error: no such file or directory: '/W0'\r\nclang: error: no such file or directory: '/Zc:__cplusplus'\r\nclang: error: no such file or directory: '/D_USE_MATH_DEFINES'\r\nclang: error: no such file or directory: '/d2ReducedOptimizeHugeFunctions'\r\nclang: error: no such file or directory: '/Zc:preprocessor'\r\nclang: error: no such file or directory: '/d2ReducedOptimizeHugeFunctions'\r\nclang: error: no such file or directory: '/std:c++17'\r\nclang: error: no such file or directory: ''--std=c++11''\r\nTarget //tensorflow/lite/java:tensorflowlite failed to build\r\nINFO: Elapsed time: 0.638s, Critical Path: 0.39s\r\nINFO: 21 processes: 21 internal.\r\nFAILED: Build did NOT complete successfully\r\n\r\nbuild command :bazel build --cxxopt='--std=c++11' //tensorflow/lite/java:tensorflowlite --crosstool_top=//external:android/crosstool --host_crosstool_top=@bazel_tools//tools/cpp:toolchain --cpu=armeabi-v7a --java_runtime_version=remotejdk_11", "@sushreebarsa hello, when I change command : bazel build -c opt --fat_apk_cpu=arm64-v8a,armeabi-v7a --host_crosstool_top=@bazel_tools//tools/cpp:toolchain //tensorflow/lite/java:tensorflow-lite --java_runtime_version=remotejdk_11, the error change to:\r\n\r\nINFO: Found applicable config definition build:short_logs in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --output_filter=DONT_MATCH_ANYTHING\r\nINFO: Found applicable config definition build:v2 in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1\r\nINFO: Found applicable config definition build:windows in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --copt=/W0 --host_copt=/W0 --copt=/Zc:__cplusplus --host_copt=/Zc:__cplusplus --copt=/D_USE_MATH_DEFINES --host_copt=/D_USE_MATH_DEFINES --features=compiler_param_file --copt=/d2ReducedOptimizeHugeFunctions --host_copt=/d2ReducedOptimizeHugeFunctions --cxxopt=/std:c++17 --host_cxxopt=/std:c++17 --config=monolithic --copt=-DWIN32_LEAN_AND_MEAN --host_copt=-DWIN32_LEAN_AND_MEAN --copt=-DNOGDI --host_copt=-DNOGDI --copt=/Zc:preprocessor --host_copt=/Zc:preprocessor --linkopt=/DEBUG --host_linkopt=/DEBUG --linkopt=/OPT:REF --host_linkopt=/OPT:REF --linkopt=/OPT:ICF --host_linkopt=/OPT:ICF --verbose_failures --features=compiler_param_file --distinct_host_configuration=false\r\nINFO: Found applicable config definition build:monolithic in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --define framework_shared_object=false --define tsl_protobuf_header_only=false --experimental_link_static_libraries_once=false\r\nINFO: Build options --cpu and --fat_apk_cpu have changed, discarding analysis cache.\r\nWARNING: Download from https://mirror.bazel.build/github.com/bazelbuild/rules_cc/archive/081771d4a0e9d7d3aa0eed2ef389fa4700dfb23e.tar.gz failed: class java.io.FileNotFoundException GET returned 404 Not Found\r\nWARNING: Download from https://storage.googleapis.com/mirror.tensorflow.org/github.com/google/XNNPACK/archive/659147817805d17c7be2d60bd7bbca7e780f9c82.zip failed: class java.io.FileNotFoundException GET returned 404 Not Found\r\nINFO: Analyzed target //tensorflow/lite/java:tensorflow-lite (3 packages loaded, 11363 targets configured).\r\nINFO: Found 1 target...\r\nERROR: C:/users/zhulei/_bazel_zhulei/qotdwtts/external/androidsdk/BUILD.bazel:13:25: Extracting interface @androidsdk//:dx_jar_import failed: missing input file 'external/androidsdk/build-tools/34.0.0/lib/dx.jar', owner: '@androidsdk//:build-tools/34.0.0/lib/dx.jar'\r\nERROR: C:/users/zhulei/_bazel_zhulei/qotdwtts/external/androidsdk/BUILD.bazel:13:25: Extracting interface @androidsdk//:dx_jar_import failed: 1 input file(s) do not exist\r\nTarget //tensorflow/lite/java:tensorflow-lite failed to build\r\nERROR: C:/users/zhulei/_bazel_zhulei/qotdwtts/external/androidsdk/BUILD.bazel:13:25 Extracting interface @androidsdk//:dx_jar_import failed: 1 input file(s) do not exist\r\nINFO: Elapsed time: 7.747s, Critical Path: 6.53s\r\nINFO: 205 processes: 183 internal, 22 local.\r\nFAILED: Build did NOT complete successfully\r\n\r\nit sames jdk/android sdk/android ndk version mismatching ,my version:\r\n jdk: java_runtime_version=remotejdk_11\r\nSET ANDROID_BUILD_TOOLS_VERSION=34.0.0\r\n SET ANDROID_NDK_API_LEVEL=26\r\n SET ANDROID_NDK_HOME=D:/project/android-ndk-r20b-windows-x86_64\r\n SET ANDROID_SDK_API_LEVEL=34\r\n SET ANDROID_SDK_HOME=D:/Android/Sdk\r\n\r\nis this ok? thank you \r\n", "@sushreebarsa when I change android sdk ,the same error:\r\nbazel build -c opt --fat_apk_cpu=arm64-v8a,armeabi-v7a --host_crosstool_top=@bazel_tools//tools/cpp:toolchain //tensorflow/lite/java:tensorflow-lite --java_runtime_version=remotejdk_11\r\n\r\nINFO: Options provided by the client:\r\n Inherited 'common' options: --isatty=1 --terminal_columns=151\r\nINFO: Reading rc options for 'build' from d:\\project\\deeplearning\\tensorflow\\.bazelrc:\r\n Inherited 'common' options: --experimental_repo_remote_exec\r\nINFO: Options provided by the client:\r\n 'build' options: --python_path=C:/ProgramData/Anaconda3/python.exe\r\nINFO: Reading rc options for 'build' from d:\\project\\deeplearning\\tensorflow\\.bazelrc:\r\n 'build' options: --define framework_shared_object=true --define tsl_protobuf_header_only=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=standalone -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --enable_platform_specific_config --define=with_xla_support=true --config=short_logs --config=v2 --define=no_aws_support=true --define=no_hdfs_support=true --experimental_cc_shared_library --experimental_link_static_libraries_once=false --incompatible_enforce_config_setting_visibility\r\nINFO: Reading rc options for 'build' from d:\\project\\deeplearning\\tensorflow\\.tf_configure.bazelrc:\r\n 'build' options: --action_env PYTHON_BIN_PATH=C:/ProgramData/Anaconda3/python.exe --action_env PYTHON_LIB_PATH=C:/ProgramData/Anaconda3/lib/site-packages --python_path=C:/ProgramData/Anaconda3/python.exe --copt=/d2ReducedOptimizeHugeFunctions --host_copt=/d2ReducedOptimizeHugeFunctions --action_env ANDROID_NDK_HOME=D:/project/android-ndk-r20b-windows-x86_64 --action_env ANDROID_NDK_API_LEVEL=26 --action_env ANDROID_BUILD_TOOLS_VERSION=31.0.0 --action_env ANDROID_SDK_API_LEVEL=31 --action_env ANDROID_SDK_HOME=D:/Android/Sdk\r\nINFO: Reading rc options for 'build' from d:\\project\\deeplearning\\tensorflow\\.bazelrc:\r\n 'build' options: --deleted_packages=tensorflow/compiler/mlir/tfrt,tensorflow/compiler/mlir/tfrt/benchmarks,tensorflow/compiler/mlir/tfrt/jit/python_binding,tensorflow/compiler/mlir/tfrt/jit/transforms,tensorflow/compiler/mlir/tfrt/python_tests,tensorflow/compiler/mlir/tfrt/tests,tensorflow/compiler/mlir/tfrt/tests/ir,tensorflow/compiler/mlir/tfrt/tests/analysis,tensorflow/compiler/mlir/tfrt/tests/jit,tensorflow/compiler/mlir/tfrt/tests/lhlo_to_tfrt,tensorflow/compiler/mlir/tfrt/tests/lhlo_to_jitrt,tensorflow/compiler/mlir/tfrt/tests/tf_to_corert,tensorflow/compiler/mlir/tfrt/tests/tf_to_tfrt_data,tensorflow/compiler/mlir/tfrt/tests/saved_model,tensorflow/compiler/mlir/tfrt/transforms/lhlo_gpu_to_tfrt_gpu,tensorflow/core/runtime_fallback,tensorflow/core/runtime_fallback/conversion,tensorflow/core/runtime_fallback/kernel,tensorflow/core/runtime_fallback/opdefs,tensorflow/core/runtime_fallback/runtime,tensorflow/core/runtime_fallback/util,tensorflow/core/tfrt/eager,tensorflow/core/tfrt/eager/backends/cpu,tensorflow/core/tfrt/eager/backends/gpu,tensorflow/core/tfrt/eager/core_runtime,tensorflow/core/tfrt/eager/cpp_tests/core_runtime,tensorflow/core/tfrt/gpu,tensorflow/core/tfrt/run_handler_thread_pool,tensorflow/core/tfrt/runtime,tensorflow/core/tfrt/saved_model,tensorflow/core/tfrt/graph_executor,tensorflow/core/tfrt/saved_model/tests,tensorflow/core/tfrt/tpu,tensorflow/core/tfrt/utils\r\nINFO: Found applicable config definition build:short_logs in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --output_filter=DONT_MATCH_ANYTHING\r\nINFO: Found applicable config definition build:v2 in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1\r\nINFO: Found applicable config definition build:windows in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --copt=/W0 --host_copt=/W0 --copt=/Zc:__cplusplus --host_copt=/Zc:__cplusplus --copt=/D_USE_MATH_DEFINES --host_copt=/D_USE_MATH_DEFINES --features=compiler_param_file --copt=/d2ReducedOptimizeHugeFunctions --host_copt=/d2ReducedOptimizeHugeFunctions --cxxopt=/std:c++17 --host_cxxopt=/std:c++17 --config=monolithic --copt=-DWIN32_LEAN_AND_MEAN --host_copt=-DWIN32_LEAN_AND_MEAN --copt=-DNOGDI --host_copt=-DNOGDI --copt=/Zc:preprocessor --host_copt=/Zc:preprocessor --linkopt=/DEBUG --host_linkopt=/DEBUG --linkopt=/OPT:REF --host_linkopt=/OPT:REF --linkopt=/OPT:ICF --host_linkopt=/OPT:ICF --verbose_failures --features=compiler_param_file --distinct_host_configuration=false\r\nINFO: Found applicable config definition build:monolithic in file d:\\project\\deeplearning\\tensorflow\\.bazelrc: --define framework_shared_object=false --define tsl_protobuf_header_only=false --experimental_link_static_libraries_once=false\r\nINFO: Analyzed target //tensorflow/lite/java:tensorflow-lite (0 packages loaded, 0 targets configured).\r\nINFO: Found 1 target...\r\nERROR: C:/users/zhulei/_bazel_zhulei/qotdwtts/external/ruy/ruy/BUILD:506:11: Compiling ruy/apply_multiplier.cc failed: (Exit 1): clang failed: error executing command\r\n cd /d C:/users/zhulei/_bazel_zhulei/qotdwtts/execroot/org_tensorflow\r\n SET ANDROID_BUILD_TOOLS_VERSION=31.0.0\r\n SET ANDROID_NDK_API_LEVEL=26\r\n SET ANDROID_NDK_HOME=D:/project/android-ndk-r20b-windows-x86_64\r\n SET ANDROID_SDK_API_LEVEL=31\r\n SET ANDROID_SDK_HOME=D:/Android/Sdk\r\n SET PATH=D:\\project\\install_soft\\msys64\\usr\\bin;D:\\project\\install_soft\\msys64\\bin;C:\\WINDOWS;C:\\WINDOWS\\System32;C:\\WINDOWS\\System32\\WindowsPowerShell\\v1.0;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.4\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.4\\libnvvp;C:\\Program Files (x86)\\VMware\\VMware Workstation\\bin\\;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.1\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.1\\libnvvp;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.3\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v11.3\\libnvvp;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v10.0\\bin;C:\\Program Files\\NVIDIA GPU Computing Toolkit\\CUDA\\v10.0\\libnvvp;C:\\windows\\system32;C:\\windows;C:\\windows\\System32\\Wbem;C:\\windows\\System32\\WindowsPowerShell\\v1.0\\;C:\\windows\\System32\\OpenSSH\\;C:\\Program Files (x86)\\NVIDIA Corporation\\PhysX\\Common;C:\\Program Files\\NVIDIA Corporation\\NVIDIA NvDLISR;C:\\Windows\\system32;C:\\Windows;C:\\Windows\\System32\\Wbem;C:\\Windows\\System32\\WindowsPowerShell\\v1.0\\;C:\\Windows\\System32\\OpenSSH\\;C:\\Program Files (x86)\\NetSarang\\Xftp 7\\;C:\\Program Files (x86)\\PuTTY\\;C:\\Users\\zhulei\\.dnx\\bin;C:\\Program Files\\Microsoft DNX\\Dnvm\\;C:\\Program Files\\Microsoft SQL Server\\130\\Tools\\Binn\\;C:\\Program Files\\Git\\cmd;C:\\ProgramData\\Anaconda3;C:\\ProgramData\\Anaconda3\\Library\\mingw-w64\\bin;C:\\ProgramData\\Anaconda3\\Library\\bin;C:\\ProgramData\\Anaconda3\\Scripts;E:\\project\\cmake-3.20.0-rc3-windows-x86_64\\bin;C:\\project\\platform-tools_r33.0.3-windows\\platform-tools;E:\\project\\protobuf-3.4.0-gmock\\install\\protobuf\\bin;C:\\Program Files\\IDM Computer Solutions\\UltraEdit;D:\\project\\TensorRT-8.4.2.4.Windows10.x86_64.cuda-11.6.cudnn8.4\\lib;C:\\Program Files\\NVIDIA Corporation\\Nsight Compute 2021.2.2\\;D:\\project\\Elasticsearch\\elasticsearch-8.1.2\\jdk;D:\\project\\Elasticsearch\\node-v17.9.1-win-x64;C:\\WINDOWS\\system32;C:\\WINDOWS;C:\\WINDOWS\\System32\\Wbem;C:\\WINDOWS\\System32\\WindowsPowerShell\\v1.0\\;C:\\WINDOWS\\System32\\OpenSSH\\;C:\\Program Files (x86)\\NetSarang\\Xshell 7\\;D:\\project\\cudnn-windows-x86_64-8.4.0.27_cuda11.6-archive\\bin;D:\\project\\install_soft;D:\\project\\install_soft\\msys64\\usr\\bin;C:\\Users\\zhulei\\AppData\\Local\\Microsoft\\WindowsApps;;C:\\Users\\zhulei\\AppData\\Local\\Programs\\Microsoft VS Code\\bin\r\n SET PWD=/proc/self/cwd\r\n SET PYTHON_BIN_PATH=C:/ProgramData/Anaconda3/python.exe\r\n SET PYTHON_LIB_PATH=C:/ProgramData/Anaconda3/lib/site-packages\r\n SET RUNFILES_MANIFEST_ONLY=1\r\n SET TF2_BEHAVIOR=1\r\n external\\androidndk\\ndk\\toolchains\\llvm\\prebuilt\\windows-x86_64\\bin\\clang -D__ANDROID_API__=26 -isystemexternal/androidndk/ndk/sysroot/usr/include/arm-linux-androideabi -target armv7-none-linux-androideabi -march=armv7-a -mfloat-abi=softfp -mfpu=vfpv3-d16 -gcc-toolchain external/androidndk/ndk/toolchains/arm-linux-androideabi-4.9/prebuilt/windows-x86_64 -fpic -no-canonical-prefixes -Wno-invalid-command-line-argument -Wno-unused-command-line-argument -funwind-tables -fstack-protector-strong -fno-addrsig -Werror=return-type -Werror=int-to-pointer-cast -Werror=pointer-to-int-cast -Werror=implicit-function-declaration -mthumb -Os -g -DNDEBUG -MD -MF bazel-out/android-armeabi-v7a-opt/bin/external/ruy/ruy/_objs/apply_multiplier/apply_multiplier.pic.d -frandom-seed=bazel-out/android-armeabi-v7a-opt/bin/external/ruy/ruy/_objs/apply_multiplier/apply_multiplier.pic.o -fPIC -iquote external/ruy -iquote bazel-out/android-armeabi-v7a-opt/bin/external/ruy /W0 /Zc:__cplusplus /D_USE_MATH_DEFINES /d2ReducedOptimizeHugeFunctions -DWIN32_LEAN_AND_MEAN -DNOGDI /Zc:preprocessor /d2ReducedOptimizeHugeFunctions /std:c++17 -mfpu=neon -O3 --sysroot=external/androidndk/ndk/platforms/android-26/arch-arm -isystem external/androidndk/ndk/sources/cxx-stl/llvm-libc++/include -isystem external/androidndk/ndk/sources/cxx-stl/llvm-libc++abi/include -isystem external/androidndk/ndk/sources/android/support/include -isystemexternal/androidndk/ndk/sysroot/usr/include -c external/ruy/ruy/apply_multiplier.cc -o bazel-out/android-armeabi-v7a-opt/bin/external/ruy/ruy/_objs/apply_multiplier/apply_multiplier.pic.o\r\n# Configuration: 9856a2746bfd1bed9df0387b5ffea58adc26b7231a0cfcd0f3bbb96c1e0074f4\r\n# Execution platform: @local_execution_config_platform//:platform\r\nclang: error: no such file or directory: '/W0'\r\nclang: error: no such file or directory: '/Zc:__cplusplus'\r\nclang: error: no such file or directory: '/D_USE_MATH_DEFINES'\r\nclang: error: no such file or directory: '/d2ReducedOptimizeHugeFunctions'\r\nclang: error: no such file or directory: '/Zc:preprocessor'\r\nclang: error: no such file or directory: '/d2ReducedOptimizeHugeFunctions'\r\nclang: error: no such file or directory: '/std:c++17'\r\nTarget //tensorflow/lite/java:tensorflow-lite failed to build\r\nINFO: Elapsed time: 1.077s, Critical Path: 0.35s\r\nINFO: 79 processes: 79 internal.\r\nFAILED: Build did NOT complete successfully\r\n\r\nis this error because:\"ERROR: C:/users/zhulei/_bazel_zhulei/qotdwtts/external/ruy/ruy/BUILD:506:11: Compiling ruy/apply_multiplier.cc failed: (Exit 1): clang failed: error executing command\r\n cd /d C:/users/zhulei/_bazel_zhulei/qotdwtts/execroot/org_tensorflow\"? but, I can not know how to sovel this problem?", "Hi @pkgoogle , please look into the issue.\r\n\r\nThank You", "Hi @leizhu1989, help me understand your goal here... is there any reason you're not following this guide? https://www.tensorflow.org/lite/android/lite_build, if not can you please follow that guide and let me know if it works for you?", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62388\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62388\">No</a>\n" ]	2023-11-14T03:55:05	2023-12-07T01:49:13	2023-12-07T01:49:10	NONE	null	null	null	System information - TensorFlow Lite version: 2.12 - OS: windows11 - build tools: bazel 5.3.0 - build command: bazel build --cxxopt='--std=c++11' //tensorflow/lite/java:tensorflowlite --crosstool_top=//external:android/crosstool --host_crosstool_top=@bazel_tools//tools/cpp:toolchain --cpu=armeabi-v7a - android NDK/SDK has installed Standalone code to reproduce the issue when I build tensorflow-lite.so for android, I cause a error: WARNING: Download from https://storage.googleapis.com/mirror.tensorflow.org/github.com/llvm/llvm-project/archive/10939d1d580b9d3c9c2f3539c6bdb39f408179c0.tar.gz failed: class java.io.FileNotFoundException GET returned 404 Not Found ERROR: C:/users/zhulei/_bazel_zhulei/qotdwtts/external/local_jdk/BUILD.bazel:2:10: in fail_rule rule @local_jdk//:jdk: Traceback (most recent call last): File "C:/users/zhulei/_bazel_zhulei/qotdwtts/external/bazel_tools/tools/jdk/fail_rule.bzl", line 19, column 13, in _fail_rule_impl fail("%s %s" % (ctx.attr.header, ctx.attr.message)) Error in fail: Auto-Configuration Error: Cannot find Java binary bin/java.exe in C:/users/zhulei/_bazel_zhulei/install/40e9c076fd7053192c88f0d437fc6512/embedded_tools/tools/jdk/nosystemjdk; either correct your JAVA_HOME, PATH or specify Java from remote repository (e.g. --java_runtime_version=remotejdk_11 ERROR: C:/users/zhulei/_bazel_zhulei/qotdwtts/external/local_jdk/BUILD.bazel:2:10: Analysis of target '@local_jdk//:jdk' failed ERROR: C:/users/zhulei/_bazel_zhulei/qotdwtts/external/bazel_tools/tools/jdk/BUILD:29:19: errors encountered resolving toolchains for @bazel_tools//tools/jdk:current_java_runtime WARNING: Download from https://golang.org/dl/?mode=json&include=all failed: class java.io.IOException connect timed out ERROR: Analysis of target '//tensorflow/lite/java:tensorflowlite' failed; build aborted: INFO: Elapsed time: 1.517s INFO: 0 processes. FAILED: Build did NOT complete successfully (0 packages loaded, 1 target configured) It sames could not find https://storage.googleapis.com/mirror.tensorflow.org/github.com/llvm/llvm-project/archive/10939d1d580b9d3c9c2f3539c6bdb39f408179c0.tar.gz , what should I do to sovel this problem,thank you !	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62388/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62388/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62387	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62387/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62387/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62387/events	https://github.com/tensorflow/tensorflow/issues/62387	1,991,865,360	I_kwDOArmXAs52uXQQ	62,387	KeyError: 'CreateRangeDecoder'	{ "login": "ZH-1225", "id": 43382527, "node_id": "MDQ6VXNlcjQzMzgyNTI3", "avatar_url": "https://avatars.githubusercontent.com/u/43382527?v=4", "gravatar_id": "", "url": "https://api.github.com/users/ZH-1225", "html_url": "https://github.com/ZH-1225", "followers_url": "https://api.github.com/users/ZH-1225/followers", "following_url": "https://api.github.com/users/ZH-1225/following{/other_user}", "gists_url": "https://api.github.com/users/ZH-1225/gists{/gist_id}", "starred_url": "https://api.github.com/users/ZH-1225/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/ZH-1225/subscriptions", "organizations_url": "https://api.github.com/users/ZH-1225/orgs", "repos_url": "https://api.github.com/users/ZH-1225/repos", "events_url": "https://api.github.com/users/ZH-1225/events{/privacy}", "received_events_url": "https://api.github.com/users/ZH-1225/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" } ]	closed	false	{ "login": "Venkat6871", "id": 147127861, "node_id": "U_kgDOCMT-NQ", "avatar_url": "https://avatars.githubusercontent.com/u/147127861?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Venkat6871", "html_url": "https://github.com/Venkat6871", "followers_url": "https://api.github.com/users/Venkat6871/followers", "following_url": "https://api.github.com/users/Venkat6871/following{/other_user}", "gists_url": "https://api.github.com/users/Venkat6871/gists{/gist_id}", "starred_url": "https://api.github.com/users/Venkat6871/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Venkat6871/subscriptions", "organizations_url": "https://api.github.com/users/Venkat6871/orgs", "repos_url": "https://api.github.com/users/Venkat6871/repos", "events_url": "https://api.github.com/users/Venkat6871/events{/privacy}", "received_events_url": "https://api.github.com/users/Venkat6871/received_events", "type": "User", "site_admin": false }	[ { "login": "Venkat6871", "id": 147127861, "node_id": "U_kgDOCMT-NQ", "avatar_url": "https://avatars.githubusercontent.com/u/147127861?v=4", "gravatar_id": "", "url": "https://api.github.com/users/Venkat6871", "html_url": "https://github.com/Venkat6871", "followers_url": "https://api.github.com/users/Venkat6871/followers", "following_url": "https://api.github.com/users/Venkat6871/following{/other_user}", "gists_url": "https://api.github.com/users/Venkat6871/gists{/gist_id}", "starred_url": "https://api.github.com/users/Venkat6871/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/Venkat6871/subscriptions", "organizations_url": "https://api.github.com/users/Venkat6871/orgs", "repos_url": "https://api.github.com/users/Venkat6871/repos", "events_url": "https://api.github.com/users/Venkat6871/events{/privacy}", "received_events_url": "https://api.github.com/users/Venkat6871/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @ZH-1225 ,\r\nCould you please fill the [template ](https://github.com/tensorflow/tensorflow/issues/new?assignees=&labels=&projects=&template=tensorflow_issue_template.yaml)which will help us to analyze the issue.\r\nThank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further." ]	2023-11-14T02:29:04	2023-12-05T01:49:22	2023-12-05T01:49:22	NONE	null	null	null	Traceback (most recent call last): File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/python/framework/ops.py", line 3022, in op_def_for_type return self._op_def_cache[type] KeyError: 'CreateRangeDecoder' During handling of the above exception, another exception occurred: Traceback (most recent call last): File "/home/xzh/.local/bin/tflite_convert", line 8, in <module> sys.exit(main()) File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/lite/python/tflite_convert.py", line 690, in main app.run(main=run_main, argv=sys.argv[:1]) File "/home/xzh/.local/lib/python3.10/site-packages/absl/app.py", line 308, in run _run_main(main, args) File "/home/xzh/.local/lib/python3.10/site-packages/absl/app.py", line 254, in _run_main sys.exit(main(argv)) File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/lite/python/tflite_convert.py", line 673, in run_main _convert_tf2_model(tflite_flags) File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/lite/python/tflite_convert.py", line 274, in _convert_tf2_model converter = lite.TFLiteConverterV2.from_saved_model( File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/lite/python/lite.py", line 2073, in from_saved_model saved_model = _load(saved_model_dir, tags) File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/python/saved_model/load.py", line 900, in load result = load_partial(export_dir, None, tags, options)["root"] File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/python/saved_model/load.py", line 1031, in load_partial loader = Loader(object_graph_proto, saved_model_proto, export_dir, File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/python/saved_model/load.py", line 161, in __init__ function_deserialization.load_function_def_library( File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/python/saved_model/function_deserialization.py", line 456, in load_function_def_library func_graph = function_def_lib.function_def_to_graph( File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/python/framework/function_def_to_graph.py", line 91, in function_def_to_graph graph_def, nested_to_flat_tensor_name = function_def_to_graph_def( File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/python/framework/function_def_to_graph.py", line 330, in function_def_to_graph_def op_def = default_graph.op_def_for_type(node_def.op) # pylint: disable=protected-access File "/home/xzh/.local/lib/python3.10/site-packages/tensorflow/python/framework/ops.py", line 3025, in op_def_for_type self._op_def_for_type(type) RuntimeError: Op type not registered 'CreateRangeDecoder' in binary running on xzh-virtual-machine. Make sure the Op and Kernel are registered in the binary running in this process. Note that if you are loading a saved graph which used ops from tf.contrib (e.g. `tf.contrib.resampler`), accessing should be done before importing the graph, as contrib ops are lazily registered when the module is first accessed.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62387/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62387/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62386	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62386/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62386/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62386/events	https://github.com/tensorflow/tensorflow/issues/62386	1,991,807,238	I_kwDOArmXAs52uJEG	62,386	[mlir][tosa] Lowering of tfl.resize_nearest_neighbor produces off-by-one error	{ "login": "sabauma", "id": 2251823, "node_id": "MDQ6VXNlcjIyNTE4MjM=", "avatar_url": "https://avatars.githubusercontent.com/u/2251823?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sabauma", "html_url": "https://github.com/sabauma", "followers_url": "https://api.github.com/users/sabauma/followers", "following_url": "https://api.github.com/users/sabauma/following{/other_user}", "gists_url": "https://api.github.com/users/sabauma/gists{/gist_id}", "starred_url": "https://api.github.com/users/sabauma/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sabauma/subscriptions", "organizations_url": "https://api.github.com/users/sabauma/orgs", "repos_url": "https://api.github.com/users/sabauma/repos", "events_url": "https://api.github.com/users/sabauma/events{/privacy}", "received_events_url": "https://api.github.com/users/sabauma/received_events", "type": "User", "site_admin": false }	[ { "id": 386191887, "node_id": "MDU6TGFiZWwzODYxOTE4ODc=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20response", "name": "stat:awaiting response", "color": "f4b400", "default": false, "description": "Status - Awaiting response from author" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 474725938, "node_id": "MDU6TGFiZWw0NzQ3MjU5Mzg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stale", "name": "stale", "color": "d4c5f9", "default": false, "description": "This label marks the issue/pr stale - to be closed automatically if no activity" }, { "id": 1097545817, "node_id": "MDU6TGFiZWwxMDk3NTQ1ODE3", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:apis", "name": "comp:apis", "color": "0052cc", "default": false, "description": "Highlevel API related issues" }, { "id": 5508003926, "node_id": "LA_kwDOArmXAs8AAAABSE14Vg", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/TF%202.13", "name": "TF 2.13", "color": "B13ACB", "default": false, "description": "For issues related to Tensorflow 2.13" } ]	closed	false	{ "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false }	[ { "login": "sushreebarsa", "id": 84765720, "node_id": "MDQ6VXNlcjg0NzY1NzIw", "avatar_url": "https://avatars.githubusercontent.com/u/84765720?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sushreebarsa", "html_url": "https://github.com/sushreebarsa", "followers_url": "https://api.github.com/users/sushreebarsa/followers", "following_url": "https://api.github.com/users/sushreebarsa/following{/other_user}", "gists_url": "https://api.github.com/users/sushreebarsa/gists{/gist_id}", "starred_url": "https://api.github.com/users/sushreebarsa/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sushreebarsa/subscriptions", "organizations_url": "https://api.github.com/users/sushreebarsa/orgs", "repos_url": "https://api.github.com/users/sushreebarsa/repos", "events_url": "https://api.github.com/users/sushreebarsa/events{/privacy}", "received_events_url": "https://api.github.com/users/sushreebarsa/received_events", "type": "User", "site_admin": false } ]	null	[ "Hello, @sabauma!\r\ntfl.resize_nearest_neighbor is [deprecated](https://colab.research.google.com/gist/sushreebarsa/1f7b8b8f65b24ad4d9ffe44a7eeca372/62386.ipynb) and not recommended to be used. Could you please try to use other up sampling operations such as tfl.upsample_bilinear and tfl.upsample_nearest instead. Generally `tf.image.resize(...method=ResizeMethod.NEAREST_NEIGHBOR...)` is referred for current use in the latest version. \r\n\r\n\r\nThank you!", "This issue is stale because it has been open for 7 days with no activity. It will be closed if no further activity occurs. Thank you.", "This issue was closed because it has been inactive for 7 days since being marked as stale. Please reopen if you'd like to work on this further.", "Are you satisfied with the resolution of your issue?\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=Yes&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62386\">Yes</a>\n<a href=\"https://docs.google.com/forms/d/e/1FAIpQLSfaP12TRhd9xSxjXZjcZFNXPGk4kc1-qMdv3gc6bEP90vY1ew/viewform?entry.85265664=No&entry.2137816233=https://github.com/tensorflow/tensorflow/issues/62386\">No</a>\n" ]	2023-11-14T01:40:56	2023-12-01T01:52:12	2023-12-01T01:52:09	CONTRIBUTOR	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? No ### Source source ### TensorFlow version 2.13 ### Custom code Yes ### OS platform and distribution Debian 11 ### Mobile device _No response_ ### Python version 3.9.2 ### Bazel version 6.1.0 ### GCC/compiler version 10.2.1 ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? Certain configurations of `tfl.resize_nearest_neighbor` produce an off-by-one error when upsampling the input tensor when using the TFL->TOSA backend. This causes the resulting image to be shifted by one pixel. ### Standalone code to reproduce the issue ```shell Below is `tfl.resize_nearest_neighbor` operation, what it lowers to using the TFL->TOSA pipeline, and a driver main function for ease of reproduction. I've included the result of lowering `tfl.resize_nearest_neighbor` in case that is easier to work with. // Original tfl.resize_nearest_neighbor operator func.func @predict(%arg0: tensor<1x4x5x3xf32>) -> (tensor<1x8x10x3xf32>) { %0 = "tfl.pseudo_const"() {value = dense<[8, 10]> : tensor<2xi32>} : () -> tensor<2xi32> %1 = "tfl.resize_nearest_neighbor"(%arg0, %0) {align_corners = false, half_pixel_centers = true} : (tensor<1x4x5x3xf32>, tensor<2xi32>) -> tensor<1x8x10x3xf32> return %1 : tensor<1x8x10x3xf32> } // After lowering to TOSA // func.func @predict(%arg0: tensor<1x4x5x3xf32>) -> tensor<1x8x10x3xf32> { // %0 = "tosa.resize"(%arg0) <{border = array<i64: 3, 3>, mode = "NEAREST_NEIGHBOR", offset = array<i64: 1, 1>, scale = array<i64: 4, 2, 4, 2>}> : (tensor<1x4x5x3xf32>) -> tensor<1x8x10x3xf32> // return %0 : tensor<1x8x10x3xf32> // } // Driver main function (for convenience) func.func @main() { %input = tensor.generate { ^bb0(%n: index, %h: index, %w: index, %c: index): %val = arith.index_cast %h : index to i64 %out = arith.uitofp %val : i64 to f32 tensor.yield %out : f32 } : tensor<1x4x5x3xf32> %out = func.call @predict(%input) : (tensor<1x4x5x3xf32>) -> (tensor<1x8x10x3xf32>) %c0 = arith.constant 0 : index %c8 = arith.constant 8 : index %c1 = arith.constant 1 : index %arg6 = tensor.cast %out : tensor<1x8x10x3xf32> to tensor<xf32> call @printMemrefF32(%arg6) : (tensor<xf32>) -> () return } func.func private @printF32(f32) func.func private @printMemrefF32(%ptr : tensor<*xf32>) The smallest Python reproducer I can produce. I am not certain if `tf.compat.v1.image.resize_nearest_neighbor` is intended to be the reference implementation of `tfl.resize_nearest_neighbor`, but it produces the output that matches the model where we are seeing the answer divergence. ```python import numpy as np import tensorflow as tf data = np.zeros([1, 4, 5, 3]) for i in range(4): data[:,i,:,:] = i expected = tf.compat.v1.image.resize_nearest_neighbor( data, [8, 10], half_pixel_centers=True, align_corners=False) print(expected) ``` Below is the output of the MLIR example on the left and the Python reference implementation on the right. The upsampling on the MLIR version replicates the value `0` only once, while the Python implementation replicates it twice. ```text tfl.resize_nearest_neighbor tf.compat.v1.image.resize_nearest_neighbor [[[[0, 0, 0], [[[[0. 0. 0.] [0, 0, 0], [0. 0. 0.] [0, 0, 0], [0. 0. 0.] [0, 0, 0], [0. 0. 0.] [0, 0, 0], [0. 0. 0.] [0, 0, 0], [0. 0. 0.] [0, 0, 0], [0. 0. 0.] [0, 0, 0], [0. 0. 0.] [0, 0, 0], [0. 0. 0.] [0, 0, 0]], [0. 0. 0.]] [[1, 1, 1], [[0. 0. 0.] [1, 1, 1], [0. 0. 0.] [1, 1, 1], [0. 0. 0.] [1, 1, 1], [0. 0. 0.] [1, 1, 1], [0. 0. 0.] [1, 1, 1], [0. 0. 0.] [1, 1, 1], [0. 0. 0.] [1, 1, 1], [0. 0. 0.] [1, 1, 1], [0. 0. 0.] [1, 1, 1]], [0. 0. 0.]] [[1, 1, 1], [[1. 1. 1.] [1, 1, 1], [1. 1. 1.] [1, 1, 1], [1. 1. 1.] [1, 1, 1], [1. 1. 1.] [1, 1, 1], [1. 1. 1.] [1, 1, 1], [1. 1. 1.] [1, 1, 1], [1. 1. 1.] [1, 1, 1], [1. 1. 1.] [1, 1, 1], [1. 1. 1.] [1, 1, 1]], [1. 1. 1.]] [[2, 2, 2], [[1. 1. 1.] [2, 2, 2], [1. 1. 1.] [2, 2, 2], [1. 1. 1.] [2, 2, 2], [1. 1. 1.] [2, 2, 2], [1. 1. 1.] [2, 2, 2], [1. 1. 1.] [2, 2, 2], [1. 1. 1.] [2, 2, 2], [1. 1. 1.] [2, 2, 2], [1. 1. 1.] [2, 2, 2]], [1. 1. 1.]] [[2, 2, 2], [[2. 2. 2.] [2, 2, 2], [2. 2. 2.] [2, 2, 2], [2. 2. 2.] [2, 2, 2], [2. 2. 2.] [2, 2, 2], [2. 2. 2.] [2, 2, 2], [2. 2. 2.] [2, 2, 2], [2. 2. 2.] [2, 2, 2], [2. 2. 2.] [2, 2, 2], [2. 2. 2.] [2, 2, 2]], [2. 2. 2.]] [[3, 3, 3], [[2. 2. 2.] [3, 3, 3], [2. 2. 2.] [3, 3, 3], [2. 2. 2.] [3, 3, 3], [2. 2. 2.] [3, 3, 3], [2. 2. 2.] [3, 3, 3], [2. 2. 2.] [3, 3, 3], [2. 2. 2.] [3, 3, 3], [2. 2. 2.] [3, 3, 3], [2. 2. 2.] [3, 3, 3]], [2. 2. 2.]] [[3, 3, 3], [[3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3]], [3. 3. 3.]] [[3, 3, 3], [[3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3], [3. 3. 3.] [3, 3, 3]]]] [3. 3. 3.]]]], shape=(1, 8, 10, 3), dtype=float64) ``` ``` ### Relevant log output _No response_	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62386/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62386/timeline	null	completed	false
https://api.github.com/repos/tensorflow/tensorflow/issues/62385	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62385/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62385/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62385/events	https://github.com/tensorflow/tensorflow/pull/62385	1,991,529,832	PR_kwDOArmXAs5fWLxZ	62,385	ConvGeneric: fix local mem reads when pointers not supported	{ "login": "looi", "id": 1855766, "node_id": "MDQ6VXNlcjE4NTU3NjY=", "avatar_url": "https://avatars.githubusercontent.com/u/1855766?v=4", "gravatar_id": "", "url": "https://api.github.com/users/looi", "html_url": "https://github.com/looi", "followers_url": "https://api.github.com/users/looi/followers", "following_url": "https://api.github.com/users/looi/following{/other_user}", "gists_url": "https://api.github.com/users/looi/gists{/gist_id}", "starred_url": "https://api.github.com/users/looi/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/looi/subscriptions", "organizations_url": "https://api.github.com/users/looi/orgs", "repos_url": "https://api.github.com/users/looi/repos", "events_url": "https://api.github.com/users/looi/events{/privacy}", "received_events_url": "https://api.github.com/users/looi/received_events", "type": "User", "site_admin": false }	[ { "id": 390482148, "node_id": "MDU6TGFiZWwzOTA0ODIxNDg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/awaiting%20review", "name": "awaiting review", "color": "bc3869", "default": false, "description": "Pull request awaiting review" }, { "id": 750616506, "node_id": "MDU6TGFiZWw3NTA2MTY1MDY=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/comp:lite", "name": "comp:lite", "color": "0052cc", "default": false, "description": "TF Lite related issues" }, { "id": 987666414, "node_id": "MDU6TGFiZWw5ODc2NjY0MTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/ready%20to%20pull", "name": "ready to pull", "color": "2cd643", "default": false, "description": "PR ready for merge process" }, { "id": 1169364259, "node_id": "MDU6TGFiZWwxMTY5MzY0MjU5", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/size:XS", "name": "size:XS", "color": "adafea", "default": false, "description": "CL Change Size: Extra Small" } ]	closed	false	{ "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false }	[ { "login": "gbaned", "id": 48215717, "node_id": "MDQ6VXNlcjQ4MjE1NzE3", "avatar_url": "https://avatars.githubusercontent.com/u/48215717?v=4", "gravatar_id": "", "url": "https://api.github.com/users/gbaned", "html_url": "https://github.com/gbaned", "followers_url": "https://api.github.com/users/gbaned/followers", "following_url": "https://api.github.com/users/gbaned/following{/other_user}", "gists_url": "https://api.github.com/users/gbaned/gists{/gist_id}", "starred_url": "https://api.github.com/users/gbaned/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/gbaned/subscriptions", "organizations_url": "https://api.github.com/users/gbaned/orgs", "repos_url": "https://api.github.com/users/gbaned/repos", "events_url": "https://api.github.com/users/gbaned/events{/privacy}", "received_events_url": "https://api.github.com/users/gbaned/received_events", "type": "User", "site_admin": false } ]	null	[ "Thanks for your pull request! It looks like this may be your first contribution to a Google open source project. Before we can look at your pull request, you'll need to sign a Contributor License Agreement (CLA).\n\nView this [failed invocation](https://github.com/tensorflow/tensorflow/pull/62385/checks?check_run_id=18642413950) of the CLA check for more information.\n\nFor the most up to date status, view the checks section at the bottom of the pull request." ]	2023-11-13T21:40:05	2023-11-15T18:23:00	2023-11-15T05:39:49	CONTRIBUTOR	null	false	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/pulls/62385", "html_url": "https://github.com/tensorflow/tensorflow/pull/62385", "diff_url": "https://github.com/tensorflow/tensorflow/pull/62385.diff", "patch_url": "https://github.com/tensorflow/tensorflow/pull/62385.patch", "merged_at": "2023-11-15T05:39:48" }	Before this change, the weights_cache is populated but then global mem read is performed, wasting the weights_cache.	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62385/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62385/timeline	null	null	true
https://api.github.com/repos/tensorflow/tensorflow/issues/62384	https://api.github.com/repos/tensorflow/tensorflow	https://api.github.com/repos/tensorflow/tensorflow/issues/62384/labels{/name}	https://api.github.com/repos/tensorflow/tensorflow/issues/62384/comments	https://api.github.com/repos/tensorflow/tensorflow/issues/62384/events	https://github.com/tensorflow/tensorflow/issues/62384	1,990,744,996	I_kwDOArmXAs52qFuk	62,384	Tensorflow (MLIR) build fails with local LLVM (Support Library cannot be built, b/c/ llvm_zlib//:zlib and llvm_zstd//:zstd are missing)	{ "login": "FabianSchuetze", "id": 10357496, "node_id": "MDQ6VXNlcjEwMzU3NDk2", "avatar_url": "https://avatars.githubusercontent.com/u/10357496?v=4", "gravatar_id": "", "url": "https://api.github.com/users/FabianSchuetze", "html_url": "https://github.com/FabianSchuetze", "followers_url": "https://api.github.com/users/FabianSchuetze/followers", "following_url": "https://api.github.com/users/FabianSchuetze/following{/other_user}", "gists_url": "https://api.github.com/users/FabianSchuetze/gists{/gist_id}", "starred_url": "https://api.github.com/users/FabianSchuetze/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/FabianSchuetze/subscriptions", "organizations_url": "https://api.github.com/users/FabianSchuetze/orgs", "repos_url": "https://api.github.com/users/FabianSchuetze/repos", "events_url": "https://api.github.com/users/FabianSchuetze/events{/privacy}", "received_events_url": "https://api.github.com/users/FabianSchuetze/received_events", "type": "User", "site_admin": false }	[ { "id": 404586594, "node_id": "MDU6TGFiZWw0MDQ1ODY1OTQ=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/stat:awaiting%20tensorflower", "name": "stat:awaiting tensorflower", "color": "f4b400", "default": false, "description": "Status - Awaiting response from tensorflower" }, { "id": 473172988, "node_id": "MDU6TGFiZWw0NzMxNzI5ODg=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:bug", "name": "type:bug", "color": "159b2e", "default": false, "description": "Bug" }, { "id": 473173351, "node_id": "MDU6TGFiZWw0NzMxNzMzNTE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:build/install", "name": "type:build/install", "color": "159b2e", "default": false, "description": "Build and install issues" }, { "id": 473184161, "node_id": "MDU6TGFiZWw0NzMxODQxNjE=", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/type:support", "name": "type:support", "color": "159b2e", "default": false, "description": "Support issues" }, { "id": 1205615612, "node_id": "MDU6TGFiZWwxMjA1NjE1NjEy", "url": "https://api.github.com/repos/tensorflow/tensorflow/labels/subtype:%20ubuntu/linux", "name": "subtype: ubuntu/linux", "color": "b619ea", "default": false, "description": "Ubuntu/Linux Build/Installation Issues" } ]	open	false	{ "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false }	[ { "login": "sachinprasadhs", "id": 73069040, "node_id": "MDQ6VXNlcjczMDY5MDQw", "avatar_url": "https://avatars.githubusercontent.com/u/73069040?v=4", "gravatar_id": "", "url": "https://api.github.com/users/sachinprasadhs", "html_url": "https://github.com/sachinprasadhs", "followers_url": "https://api.github.com/users/sachinprasadhs/followers", "following_url": "https://api.github.com/users/sachinprasadhs/following{/other_user}", "gists_url": "https://api.github.com/users/sachinprasadhs/gists{/gist_id}", "starred_url": "https://api.github.com/users/sachinprasadhs/starred{/owner}{/repo}", "subscriptions_url": "https://api.github.com/users/sachinprasadhs/subscriptions", "organizations_url": "https://api.github.com/users/sachinprasadhs/orgs", "repos_url": "https://api.github.com/users/sachinprasadhs/repos", "events_url": "https://api.github.com/users/sachinprasadhs/events{/privacy}", "received_events_url": "https://api.github.com/users/sachinprasadhs/received_events", "type": "User", "site_admin": false } ]	null	[ "Hi @FabianSchuetze ,\r\n\r\nI think as per documentation the command is which is slight change wrt target.\r\n\r\n```\r\nbazel build --override_repository=\"llvm-raw=${LLVM_SRC}\" \\\r\n -c opt tensorflow/compiler/mlir:tf-opt\r\n```\r\n\r\nAs per Bazel [documentation](https://bazel.build/run/build#specifying-build-targets), all target patterns starting with `//` are resolved relative to the current workspace.Target patterns that do not begin with `//` are resolved relative to the current working directory. \r\n\r\nIs that making difference here ? Could you try with exact command mentioned in documentation?\r\n\r\n", "Thanks for your reply, @SuryanarayanaY .\r\n\r\nYour proposed solution did not help. In fact, the error is exactly as before:\r\n\r\n```\r\nERROR: /home/fasc6540/.cache/bazel/_bazel_fasc6540/e5cce820cc082410b4fcc604db349066/external/llvm-project/llvm/BUILD.bazel:195:11: no such package '@llvm_zlib//': The repository '@llvm_zlib' could not be resolved: Repository '@llvm_zlib' is not defined and referenced by '@llvm-project//llvm:Support'\r\n```\r\n\r\nThe problem is that it is not clear how to build @llvm_zlib//:zlib, and @llvm_zstd//:zstd cannot be build but are referenced by the `Support` library. ", "CC: @learning-to-play ", "I have been able to build tf-opt by removing the local llvm. bazel build -c opt //tensorflow/compiler/mlir:tf-opt", "> bazel build -c opt //tensorflow/compiler/mlir:tf-opt\r\n\r\nThis didn't work for me, I am getting this error.\r\n\r\n```bash\r\nERROR: /tensorflow_src/tensorflow/core/common_runtime/BUILD:3386:11: Compiling tensorflow/core/common_runtime/optimized_function_graph_info.cc failed: (Exit 1): gcc failed: error executing command (from target //tensorflow/core/common_runtime:optimized_function_graph_info)\r\n```", "@FabianSchuetze Hi did you solve this problem?", "This could be a possible workaround?\r\n\r\nhttps://github.com/tensorflow/tensorflow/issues/69367#issuecomment-2155693927" ]	2023-11-13T14:13:18	2024-06-07T23:33:37	null	NONE	null	null	null	### Issue type Bug ### Have you reproduced the bug with TensorFlow Nightly? Yes ### Source source ### TensorFlow version 01b930daff1e6437f813a2ceb2ee7bdb174babd3 ### Custom code Yes ### OS platform and distribution Ubuntu 222.04 ### Mobile device _No response_ ### Python version _No response_ ### Bazel version 6.1.0 ### GCC/compiler version 11.4.0 ### CUDA/cuDNN version _No response_ ### GPU model and memory _No response_ ### Current behavior? I am following the [instructions](https://github.com/tensorflow/tensorflow/blob/master/tensorflow/compiler/mlir/README.md) to build `mlir:tf-opt-opt` with a local LLVM copy and get have the following error: ``` ➜ tensorflow git:(master) bazel build --override_repository="llvm-raw=${LLVM_SRC}" --config=dbg //tensorflow/compiler/mlir:tf-opt INFO: Reading 'startup' options from /tmp/tensorflow/.bazelrc: --windows_enable_symlinks INFO: Options provided by the client: Inherited 'common' options: --isatty=1 --terminal_columns=104 INFO: Reading rc options for 'build' from /tmp/tensorflow/.bazelrc: Inherited 'common' options: --experimental_repo_remote_exec INFO: Reading rc options for 'build' from /tmp/tensorflow/.bazelrc: 'build' options: --define framework_shared_object=true --define tsl_protobuf_header_only=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=standalone -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --features=-force_no_whole_archive --enable_platform_specific_config --define=with_xla_support=true --config=short_logs --config=v2 --define=no_aws_support=true --define=no_hdfs_support=true --experimental_cc_shared_library --experimental_link_static_libraries_once=false --incompatible_enforce_config_setting_visibility INFO: Found applicable config definition build:short_logs in file /tmp/tensorflow/.bazelrc: --output_filter=DONT_MATCH_ANYTHING INFO: Found applicable config definition build:v2 in file /tmp/tensorflow/.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1 INFO: Found applicable config definition build:dbg in file /tmp/tensorflow/.bazelrc: -c dbg --per_file_copt=+.,-tensorflow.,-xla.@-g0 --per_file_copt=+tensorflow/core/kernels.@-g0 --cxxopt -DTF_LITE_DISABLE_X86_NEON --copt -DDEBUG_BUILD INFO: Found applicable config definition build:linux in file /tmp/tensorflow/.bazelrc: --host_copt=-w --copt=-Wno-all --copt=-Wno-extra --copt=-Wno-deprecated --copt=-Wno-deprecated-declarations --copt=-Wno-ignored-attributes --copt=-Wno-array-bounds --copt=-Wunused-result --copt=-Werror=unused-result --copt=-Wswitch --copt=-Werror=switch --copt=-Wno-error=unused-but-set-variable --define=PREFIX=/usr --define=LIBDIR=$(PREFIX)/lib --define=INCLUDEDIR=$(PREFIX)/include --define=PROTOBUF_INCLUDE_PATH=$(PREFIX)/include --cxxopt=-std=c++17 --host_cxxopt=-std=c++17 --config=dynamic_kernels --experimental_guard_against_concurrent_changes INFO: Found applicable config definition build:dynamic_kernels in file /tmp/tensorflow/.bazelrc: --define=dynamic_loaded_kernels=true --copt=-DAUTOLOAD_DYNAMIC_KERNELS INFO: Repository com_googlesource_code_re2 instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:918:21: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:266:20: in _tf_repositories /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> INFO: Repository gif instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:918:21: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:331:20: in _tf_repositories /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> INFO: Repository stablehlo instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:911:28: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:88:14: in _initialize_third_party /tmp/tensorflow/third_party/stablehlo/workspace.bzl:11:20: in repo /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> INFO: Repository onednn instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:918:21: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:197:20: in _tf_repositories /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> INFO: Repository riegeli instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:918:21: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:848:20: in _tf_repositories /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> ERROR: /home/fasc6540/.cache/bazel/_bazel_fasc6540/e5cce820cc082410b4fcc604db349066/external/llvm-project/llvm/BUILD.bazel:195:11: no such package '@llvm_zlib//': The repository '@llvm_zlib' could not be resolved: Repository '@llvm_zlib' is not defined and referenced by '@llvm-project//llvm:Support' ERROR: Analysis of target '//tensorflow/compiler/mlir:tf-opt' failed; build aborted: INFO: Elapsed time: 1.651s INFO: 0 processes. FAILED: Build did NOT complete successfully (171 packages loaded, 5167 targets configured) currently loading: tensorflow/compiler/mlir/quantization/tensorflow/calibrator ... (3 packages) Fetching repository @ml_dtypes; starting Fetching repository @double_conversion; starting Fetching repository @snappy; starting Fetching .../e5cce820cc082410b4fcc604db349066/external/double_conversion; Extracting v3.2.0.tar.gz Fetching ...04db349066/external/snappy; Extracting 984b191f0fefdeb17050b42a90b7625999c13b8d.tar.gz Fetching repository @zlib; starting Fetching repository @jsoncpp_git; starting Fetching ...cc082410b4fcc604db349066/external/zlib; Extracting zlib-1.2.13.tar.gz ... (16 fetches) ``` Is there anything I can do to address the issue? ### Standalone code to reproduce the issue ```shell The steps I did to reproduce the issue are: cd /tmp/ git clone https://github.com/tensorflow/tensorflow.git git clone https://github.com/llvm/llvm-project.git export LLVM_SRC=/tmp/llvm-project/ touch ${LLVM_SRC}/BUILD.bazel ${LLVM_SRC}/WORKSPACE bazel build --override_repository="llvm-raw=${LLVM_SRC}" --config=dbg //tensorflow/compiler/mlir:tf-opt ``` ``` ### Relevant log output ```shell ➜ tensorflow git:(master) bazel build --override_repository="llvm-raw=${LLVM_SRC}" --config=dbg //tensorflow/compiler/mlir:tf-opt INFO: Reading 'startup' options from /tmp/tensorflow/.bazelrc: --windows_enable_symlinks INFO: Options provided by the client: Inherited 'common' options: --isatty=1 --terminal_columns=104 INFO: Reading rc options for 'build' from /tmp/tensorflow/.bazelrc: Inherited 'common' options: --experimental_repo_remote_exec INFO: Reading rc options for 'build' from /tmp/tensorflow/.bazelrc: 'build' options: --define framework_shared_object=true --define tsl_protobuf_header_only=true --define=use_fast_cpp_protos=true --define=allow_oversize_protos=true --spawn_strategy=standalone -c opt --announce_rc --define=grpc_no_ares=true --noincompatible_remove_legacy_whole_archive --features=-force_no_whole_archive --enable_platform_specific_config --define=with_xla_support=true --config=short_logs --config=v2 --define=no_aws_support=true --define=no_hdfs_support=true --experimental_cc_shared_library --experimental_link_static_libraries_once=false --incompatible_enforce_config_setting_visibility INFO: Found applicable config definition build:short_logs in file /tmp/tensorflow/.bazelrc: --output_filter=DONT_MATCH_ANYTHING INFO: Found applicable config definition build:v2 in file /tmp/tensorflow/.bazelrc: --define=tf_api_version=2 --action_env=TF2_BEHAVIOR=1 INFO: Found applicable config definition build:dbg in file /tmp/tensorflow/.bazelrc: -c dbg --per_file_copt=+.,-tensorflow.,-xla.@-g0 --per_file_copt=+tensorflow/core/kernels.@-g0 --cxxopt -DTF_LITE_DISABLE_X86_NEON --copt -DDEBUG_BUILD INFO: Found applicable config definition build:linux in file /tmp/tensorflow/.bazelrc: --host_copt=-w --copt=-Wno-all --copt=-Wno-extra --copt=-Wno-deprecated --copt=-Wno-deprecated-declarations --copt=-Wno-ignored-attributes --copt=-Wno-array-bounds --copt=-Wunused-result --copt=-Werror=unused-result --copt=-Wswitch --copt=-Werror=switch --copt=-Wno-error=unused-but-set-variable --define=PREFIX=/usr --define=LIBDIR=$(PREFIX)/lib --define=INCLUDEDIR=$(PREFIX)/include --define=PROTOBUF_INCLUDE_PATH=$(PREFIX)/include --cxxopt=-std=c++17 --host_cxxopt=-std=c++17 --config=dynamic_kernels --experimental_guard_against_concurrent_changes INFO: Found applicable config definition build:dynamic_kernels in file /tmp/tensorflow/.bazelrc: --define=dynamic_loaded_kernels=true --copt=-DAUTOLOAD_DYNAMIC_KERNELS INFO: Repository com_googlesource_code_re2 instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:918:21: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:266:20: in _tf_repositories /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> INFO: Repository gif instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:918:21: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:331:20: in _tf_repositories /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> INFO: Repository stablehlo instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:911:28: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:88:14: in _initialize_third_party /tmp/tensorflow/third_party/stablehlo/workspace.bzl:11:20: in repo /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> INFO: Repository onednn instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:918:21: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:197:20: in _tf_repositories /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> INFO: Repository riegeli instantiated at: /tmp/tensorflow/WORKSPACE:84:14: in <toplevel> /tmp/tensorflow/tensorflow/workspace2.bzl:918:21: in workspace /tmp/tensorflow/tensorflow/workspace2.bzl:848:20: in _tf_repositories /tmp/tensorflow/third_party/repo.bzl:136:21: in tf_http_archive Repository rule _tf_http_archive defined at: /tmp/tensorflow/third_party/repo.bzl:89:35: in <toplevel> ERROR: /home/fasc6540/.cache/bazel/_bazel_fasc6540/e5cce820cc082410b4fcc604db349066/external/llvm-project/llvm/BUILD.bazel:195:11: no such package '@llvm_zlib//': The repository '@llvm_zlib' could not be resolved: Repository '@llvm_zlib' is not defined and referenced by '@llvm-project//llvm:Support' ERROR: Analysis of target '//tensorflow/compiler/mlir:tf-opt' failed; build aborted: INFO: Elapsed time: 1.651s INFO: 0 processes. FAILED: Build did NOT complete successfully (171 packages loaded, 5167 targets configured) currently loading: tensorflow/compiler/mlir/quantization/tensorflow/calibrator ... (3 packages) Fetching repository @ml_dtypes; starting Fetching repository @double_conversion; starting Fetching repository @snappy; starting Fetching .../e5cce820cc082410b4fcc604db349066/external/double_conversion; Extracting v3.2.0.tar.gz Fetching ...04db349066/external/snappy; Extracting 984b191f0fefdeb17050b42a90b7625999c13b8d.tar.gz Fetching repository @zlib; starting Fetching repository @jsoncpp_git; starting Fetching ...cc082410b4fcc604db349066/external/zlib; Extracting zlib-1.2.13.tar.gz ... (16 fetches) ```	{ "url": "https://api.github.com/repos/tensorflow/tensorflow/issues/62384/reactions", "total_count": 0, "+1": 0, "-1": 0, "laugh": 0, "hooray": 0, "confused": 0, "heart": 0, "rocket": 0, "eyes": 0 }	https://api.github.com/repos/tensorflow/tensorflow/issues/62384/timeline	null	null	false

Subsets and Splits

No community queries yet

The top public SQL queries from the community will appear here once available.