Spaces:
Sleeping
Sleeping
Upload 10 files
Browse files- app.py +30 -0
- checkpoint-600/config.json +53 -0
- checkpoint-600/model.safetensors +3 -0
- checkpoint-600/optimizer.pt +3 -0
- checkpoint-600/preprocessor_config.json +36 -0
- checkpoint-600/rng_state.pth +3 -0
- checkpoint-600/scheduler.pt +3 -0
- checkpoint-600/trainer_state.json +486 -0
- checkpoint-600/training_args.bin +3 -0
- requirements.txt +4 -0
app.py
ADDED
|
@@ -0,0 +1,30 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
import torch
|
| 2 |
+
from transformers import pipeline
|
| 3 |
+
import torch
|
| 4 |
+
|
| 5 |
+
import gradio as gr
|
| 6 |
+
|
| 7 |
+
with gr.Blocks() as demo:
|
| 8 |
+
def submit(image):
|
| 9 |
+
torch.backends.cuda.matmul.allow_tf32 = True
|
| 10 |
+
torch.backends.cudnn.allow_tf32 = True
|
| 11 |
+
pipe = pipeline("image-classification", model=".\\checkpoint-600")
|
| 12 |
+
output = pipe(images=[image])
|
| 13 |
+
|
| 14 |
+
result = {}
|
| 15 |
+
|
| 16 |
+
for index, item in enumerate(output[0]):
|
| 17 |
+
result[item["label"]] = item["score"]
|
| 18 |
+
|
| 19 |
+
return result
|
| 20 |
+
|
| 21 |
+
with gr.Row():
|
| 22 |
+
with gr.Column():
|
| 23 |
+
image_input = gr.Image(label="Input image", type="filepath")
|
| 24 |
+
|
| 25 |
+
with gr.Column():
|
| 26 |
+
label = gr.Label()
|
| 27 |
+
submit_button = gr.Button(value="Submit", variant="primary")
|
| 28 |
+
submit_button.click(submit, inputs=[image_input], outputs=label)
|
| 29 |
+
|
| 30 |
+
demo.launch()
|
checkpoint-600/config.json
ADDED
|
@@ -0,0 +1,53 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_name_or_path": "google/vit-base-patch16-224-in21k",
|
| 3 |
+
"architectures": [
|
| 4 |
+
"ViTForImageClassification"
|
| 5 |
+
],
|
| 6 |
+
"attention_probs_dropout_prob": 0.0,
|
| 7 |
+
"encoder_stride": 16,
|
| 8 |
+
"finetuning_task": "image-classification",
|
| 9 |
+
"hidden_act": "gelu",
|
| 10 |
+
"hidden_dropout_prob": 0.0,
|
| 11 |
+
"hidden_size": 768,
|
| 12 |
+
"id2label": {
|
| 13 |
+
"0": "blured_man",
|
| 14 |
+
"1": "deanon_modov",
|
| 15 |
+
"10": "surprised_face",
|
| 16 |
+
"11": "ua_flag",
|
| 17 |
+
"2": "money",
|
| 18 |
+
"3": "nsfw",
|
| 19 |
+
"4": "ok",
|
| 20 |
+
"5": "shinobu",
|
| 21 |
+
"6": "soldatik_opt1",
|
| 22 |
+
"7": "soldatik_opt2",
|
| 23 |
+
"8": "soldatik_opt3",
|
| 24 |
+
"9": "sosok"
|
| 25 |
+
},
|
| 26 |
+
"image_size": 224,
|
| 27 |
+
"initializer_range": 0.02,
|
| 28 |
+
"intermediate_size": 3072,
|
| 29 |
+
"label2id": {
|
| 30 |
+
"blured_man": "0",
|
| 31 |
+
"deanon_modov": "1",
|
| 32 |
+
"money": "2",
|
| 33 |
+
"nsfw": "3",
|
| 34 |
+
"ok": "4",
|
| 35 |
+
"shinobu": "5",
|
| 36 |
+
"soldatik_opt1": "6",
|
| 37 |
+
"soldatik_opt2": "7",
|
| 38 |
+
"soldatik_opt3": "8",
|
| 39 |
+
"sosok": "9",
|
| 40 |
+
"surprised_face": "10",
|
| 41 |
+
"ua_flag": "11"
|
| 42 |
+
},
|
| 43 |
+
"layer_norm_eps": 1e-12,
|
| 44 |
+
"model_type": "vit",
|
| 45 |
+
"num_attention_heads": 12,
|
| 46 |
+
"num_channels": 3,
|
| 47 |
+
"num_hidden_layers": 12,
|
| 48 |
+
"patch_size": 16,
|
| 49 |
+
"problem_type": "single_label_classification",
|
| 50 |
+
"qkv_bias": true,
|
| 51 |
+
"torch_dtype": "float32",
|
| 52 |
+
"transformers_version": "4.39.0.dev0"
|
| 53 |
+
}
|
checkpoint-600/model.safetensors
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:a15084478cfe7b13e29f6f2fbe844dea802753bce0d1cd2a15c22bca3b411be3
|
| 3 |
+
size 343254736
|
checkpoint-600/optimizer.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:e25d3ed4c78c2cfbf1da6017eafa10a2948fdc86c2aa0ecaf7872d1917ec7150
|
| 3 |
+
size 686630330
|
checkpoint-600/preprocessor_config.json
ADDED
|
@@ -0,0 +1,36 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"_valid_processor_keys": [
|
| 3 |
+
"images",
|
| 4 |
+
"do_resize",
|
| 5 |
+
"size",
|
| 6 |
+
"resample",
|
| 7 |
+
"do_rescale",
|
| 8 |
+
"rescale_factor",
|
| 9 |
+
"do_normalize",
|
| 10 |
+
"image_mean",
|
| 11 |
+
"image_std",
|
| 12 |
+
"return_tensors",
|
| 13 |
+
"data_format",
|
| 14 |
+
"input_data_format"
|
| 15 |
+
],
|
| 16 |
+
"do_normalize": true,
|
| 17 |
+
"do_rescale": true,
|
| 18 |
+
"do_resize": true,
|
| 19 |
+
"image_mean": [
|
| 20 |
+
0.5,
|
| 21 |
+
0.5,
|
| 22 |
+
0.5
|
| 23 |
+
],
|
| 24 |
+
"image_processor_type": "ViTImageProcessor",
|
| 25 |
+
"image_std": [
|
| 26 |
+
0.5,
|
| 27 |
+
0.5,
|
| 28 |
+
0.5
|
| 29 |
+
],
|
| 30 |
+
"resample": 2,
|
| 31 |
+
"rescale_factor": 0.00392156862745098,
|
| 32 |
+
"size": {
|
| 33 |
+
"height": 224,
|
| 34 |
+
"width": 224
|
| 35 |
+
}
|
| 36 |
+
}
|
checkpoint-600/rng_state.pth
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:4499333e86e13d31c7647ccfc8ed8f90b214a0035f17a87dab71ad7caf22d3cc
|
| 3 |
+
size 14308
|
checkpoint-600/scheduler.pt
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:13f3997a48c6c1f3b91244cb7fd4fdd615afc9185f3b6fa9544d96c0c0ef490d
|
| 3 |
+
size 1064
|
checkpoint-600/trainer_state.json
ADDED
|
@@ -0,0 +1,486 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
{
|
| 2 |
+
"best_metric": 0.2971116304397583,
|
| 3 |
+
"best_model_checkpoint": "./svin/checkpoint-600",
|
| 4 |
+
"epoch": 5.0,
|
| 5 |
+
"eval_steps": 500,
|
| 6 |
+
"global_step": 600,
|
| 7 |
+
"is_hyper_param_search": false,
|
| 8 |
+
"is_local_process_zero": true,
|
| 9 |
+
"is_world_process_zero": true,
|
| 10 |
+
"log_history": [
|
| 11 |
+
{
|
| 12 |
+
"epoch": 0.08,
|
| 13 |
+
"grad_norm": 1.4835056066513062,
|
| 14 |
+
"learning_rate": 1.9666666666666666e-05,
|
| 15 |
+
"loss": 2.4036,
|
| 16 |
+
"step": 10
|
| 17 |
+
},
|
| 18 |
+
{
|
| 19 |
+
"epoch": 0.17,
|
| 20 |
+
"grad_norm": 1.6403590440750122,
|
| 21 |
+
"learning_rate": 1.9333333333333333e-05,
|
| 22 |
+
"loss": 2.126,
|
| 23 |
+
"step": 20
|
| 24 |
+
},
|
| 25 |
+
{
|
| 26 |
+
"epoch": 0.25,
|
| 27 |
+
"grad_norm": 1.8593930006027222,
|
| 28 |
+
"learning_rate": 1.9e-05,
|
| 29 |
+
"loss": 1.7892,
|
| 30 |
+
"step": 30
|
| 31 |
+
},
|
| 32 |
+
{
|
| 33 |
+
"epoch": 0.33,
|
| 34 |
+
"grad_norm": 1.6115764379501343,
|
| 35 |
+
"learning_rate": 1.866666666666667e-05,
|
| 36 |
+
"loss": 1.5352,
|
| 37 |
+
"step": 40
|
| 38 |
+
},
|
| 39 |
+
{
|
| 40 |
+
"epoch": 0.42,
|
| 41 |
+
"grad_norm": 1.3383398056030273,
|
| 42 |
+
"learning_rate": 1.8333333333333333e-05,
|
| 43 |
+
"loss": 1.3303,
|
| 44 |
+
"step": 50
|
| 45 |
+
},
|
| 46 |
+
{
|
| 47 |
+
"epoch": 0.5,
|
| 48 |
+
"grad_norm": 1.3053131103515625,
|
| 49 |
+
"learning_rate": 1.8e-05,
|
| 50 |
+
"loss": 1.1755,
|
| 51 |
+
"step": 60
|
| 52 |
+
},
|
| 53 |
+
{
|
| 54 |
+
"epoch": 0.58,
|
| 55 |
+
"grad_norm": 1.4148197174072266,
|
| 56 |
+
"learning_rate": 1.7666666666666668e-05,
|
| 57 |
+
"loss": 1.0469,
|
| 58 |
+
"step": 70
|
| 59 |
+
},
|
| 60 |
+
{
|
| 61 |
+
"epoch": 0.67,
|
| 62 |
+
"grad_norm": 1.3164098262786865,
|
| 63 |
+
"learning_rate": 1.7333333333333336e-05,
|
| 64 |
+
"loss": 0.9351,
|
| 65 |
+
"step": 80
|
| 66 |
+
},
|
| 67 |
+
{
|
| 68 |
+
"epoch": 0.75,
|
| 69 |
+
"grad_norm": 1.351128339767456,
|
| 70 |
+
"learning_rate": 1.7e-05,
|
| 71 |
+
"loss": 0.8411,
|
| 72 |
+
"step": 90
|
| 73 |
+
},
|
| 74 |
+
{
|
| 75 |
+
"epoch": 0.83,
|
| 76 |
+
"grad_norm": 1.1163653135299683,
|
| 77 |
+
"learning_rate": 1.6666666666666667e-05,
|
| 78 |
+
"loss": 0.82,
|
| 79 |
+
"step": 100
|
| 80 |
+
},
|
| 81 |
+
{
|
| 82 |
+
"epoch": 0.92,
|
| 83 |
+
"grad_norm": 1.2481478452682495,
|
| 84 |
+
"learning_rate": 1.6333333333333335e-05,
|
| 85 |
+
"loss": 0.7225,
|
| 86 |
+
"step": 110
|
| 87 |
+
},
|
| 88 |
+
{
|
| 89 |
+
"epoch": 1.0,
|
| 90 |
+
"grad_norm": 1.494071364402771,
|
| 91 |
+
"learning_rate": 1.6000000000000003e-05,
|
| 92 |
+
"loss": 0.6972,
|
| 93 |
+
"step": 120
|
| 94 |
+
},
|
| 95 |
+
{
|
| 96 |
+
"epoch": 1.0,
|
| 97 |
+
"eval_accuracy": 0.8978622327790974,
|
| 98 |
+
"eval_loss": 0.6513156294822693,
|
| 99 |
+
"eval_runtime": 31.845,
|
| 100 |
+
"eval_samples_per_second": 92.542,
|
| 101 |
+
"eval_steps_per_second": 0.691,
|
| 102 |
+
"step": 120
|
| 103 |
+
},
|
| 104 |
+
{
|
| 105 |
+
"epoch": 1.08,
|
| 106 |
+
"grad_norm": 1.2568280696868896,
|
| 107 |
+
"learning_rate": 1.5666666666666667e-05,
|
| 108 |
+
"loss": 0.6147,
|
| 109 |
+
"step": 130
|
| 110 |
+
},
|
| 111 |
+
{
|
| 112 |
+
"epoch": 1.17,
|
| 113 |
+
"grad_norm": 1.1234235763549805,
|
| 114 |
+
"learning_rate": 1.5333333333333334e-05,
|
| 115 |
+
"loss": 0.5977,
|
| 116 |
+
"step": 140
|
| 117 |
+
},
|
| 118 |
+
{
|
| 119 |
+
"epoch": 1.25,
|
| 120 |
+
"grad_norm": 1.614490270614624,
|
| 121 |
+
"learning_rate": 1.5000000000000002e-05,
|
| 122 |
+
"loss": 0.6324,
|
| 123 |
+
"step": 150
|
| 124 |
+
},
|
| 125 |
+
{
|
| 126 |
+
"epoch": 1.33,
|
| 127 |
+
"grad_norm": 1.1563916206359863,
|
| 128 |
+
"learning_rate": 1.4666666666666666e-05,
|
| 129 |
+
"loss": 0.5423,
|
| 130 |
+
"step": 160
|
| 131 |
+
},
|
| 132 |
+
{
|
| 133 |
+
"epoch": 1.42,
|
| 134 |
+
"grad_norm": 1.1219310760498047,
|
| 135 |
+
"learning_rate": 1.4333333333333334e-05,
|
| 136 |
+
"loss": 0.5335,
|
| 137 |
+
"step": 170
|
| 138 |
+
},
|
| 139 |
+
{
|
| 140 |
+
"epoch": 1.5,
|
| 141 |
+
"grad_norm": 1.0543292760849,
|
| 142 |
+
"learning_rate": 1.4e-05,
|
| 143 |
+
"loss": 0.5169,
|
| 144 |
+
"step": 180
|
| 145 |
+
},
|
| 146 |
+
{
|
| 147 |
+
"epoch": 1.58,
|
| 148 |
+
"grad_norm": 1.5661933422088623,
|
| 149 |
+
"learning_rate": 1.3666666666666667e-05,
|
| 150 |
+
"loss": 0.5247,
|
| 151 |
+
"step": 190
|
| 152 |
+
},
|
| 153 |
+
{
|
| 154 |
+
"epoch": 1.67,
|
| 155 |
+
"grad_norm": 1.4204506874084473,
|
| 156 |
+
"learning_rate": 1.3333333333333333e-05,
|
| 157 |
+
"loss": 0.4767,
|
| 158 |
+
"step": 200
|
| 159 |
+
},
|
| 160 |
+
{
|
| 161 |
+
"epoch": 1.75,
|
| 162 |
+
"grad_norm": 1.054136872291565,
|
| 163 |
+
"learning_rate": 1.3000000000000001e-05,
|
| 164 |
+
"loss": 0.4874,
|
| 165 |
+
"step": 210
|
| 166 |
+
},
|
| 167 |
+
{
|
| 168 |
+
"epoch": 1.83,
|
| 169 |
+
"grad_norm": 1.1543879508972168,
|
| 170 |
+
"learning_rate": 1.2666666666666667e-05,
|
| 171 |
+
"loss": 0.4546,
|
| 172 |
+
"step": 220
|
| 173 |
+
},
|
| 174 |
+
{
|
| 175 |
+
"epoch": 1.92,
|
| 176 |
+
"grad_norm": 1.2403457164764404,
|
| 177 |
+
"learning_rate": 1.2333333333333334e-05,
|
| 178 |
+
"loss": 0.4272,
|
| 179 |
+
"step": 230
|
| 180 |
+
},
|
| 181 |
+
{
|
| 182 |
+
"epoch": 2.0,
|
| 183 |
+
"grad_norm": 1.166118860244751,
|
| 184 |
+
"learning_rate": 1.2e-05,
|
| 185 |
+
"loss": 0.4305,
|
| 186 |
+
"step": 240
|
| 187 |
+
},
|
| 188 |
+
{
|
| 189 |
+
"epoch": 2.0,
|
| 190 |
+
"eval_accuracy": 0.9477434679334917,
|
| 191 |
+
"eval_loss": 0.41236403584480286,
|
| 192 |
+
"eval_runtime": 31.083,
|
| 193 |
+
"eval_samples_per_second": 94.811,
|
| 194 |
+
"eval_steps_per_second": 0.708,
|
| 195 |
+
"step": 240
|
| 196 |
+
},
|
| 197 |
+
{
|
| 198 |
+
"epoch": 2.08,
|
| 199 |
+
"grad_norm": 1.3920248746871948,
|
| 200 |
+
"learning_rate": 1.1666666666666668e-05,
|
| 201 |
+
"loss": 0.3951,
|
| 202 |
+
"step": 250
|
| 203 |
+
},
|
| 204 |
+
{
|
| 205 |
+
"epoch": 2.17,
|
| 206 |
+
"grad_norm": 1.9847582578659058,
|
| 207 |
+
"learning_rate": 1.1333333333333334e-05,
|
| 208 |
+
"loss": 0.4493,
|
| 209 |
+
"step": 260
|
| 210 |
+
},
|
| 211 |
+
{
|
| 212 |
+
"epoch": 2.25,
|
| 213 |
+
"grad_norm": 0.9650905132293701,
|
| 214 |
+
"learning_rate": 1.1000000000000001e-05,
|
| 215 |
+
"loss": 0.3995,
|
| 216 |
+
"step": 270
|
| 217 |
+
},
|
| 218 |
+
{
|
| 219 |
+
"epoch": 2.33,
|
| 220 |
+
"grad_norm": 1.1868102550506592,
|
| 221 |
+
"learning_rate": 1.0666666666666667e-05,
|
| 222 |
+
"loss": 0.401,
|
| 223 |
+
"step": 280
|
| 224 |
+
},
|
| 225 |
+
{
|
| 226 |
+
"epoch": 2.42,
|
| 227 |
+
"grad_norm": 0.9317110776901245,
|
| 228 |
+
"learning_rate": 1.0333333333333335e-05,
|
| 229 |
+
"loss": 0.3785,
|
| 230 |
+
"step": 290
|
| 231 |
+
},
|
| 232 |
+
{
|
| 233 |
+
"epoch": 2.5,
|
| 234 |
+
"grad_norm": 1.3572497367858887,
|
| 235 |
+
"learning_rate": 1e-05,
|
| 236 |
+
"loss": 0.3785,
|
| 237 |
+
"step": 300
|
| 238 |
+
},
|
| 239 |
+
{
|
| 240 |
+
"epoch": 2.58,
|
| 241 |
+
"grad_norm": 0.971365749835968,
|
| 242 |
+
"learning_rate": 9.666666666666667e-06,
|
| 243 |
+
"loss": 0.3789,
|
| 244 |
+
"step": 310
|
| 245 |
+
},
|
| 246 |
+
{
|
| 247 |
+
"epoch": 2.67,
|
| 248 |
+
"grad_norm": 1.131386399269104,
|
| 249 |
+
"learning_rate": 9.333333333333334e-06,
|
| 250 |
+
"loss": 0.3636,
|
| 251 |
+
"step": 320
|
| 252 |
+
},
|
| 253 |
+
{
|
| 254 |
+
"epoch": 2.75,
|
| 255 |
+
"grad_norm": 1.010298728942871,
|
| 256 |
+
"learning_rate": 9e-06,
|
| 257 |
+
"loss": 0.3787,
|
| 258 |
+
"step": 330
|
| 259 |
+
},
|
| 260 |
+
{
|
| 261 |
+
"epoch": 2.83,
|
| 262 |
+
"grad_norm": 1.622625470161438,
|
| 263 |
+
"learning_rate": 8.666666666666668e-06,
|
| 264 |
+
"loss": 0.3829,
|
| 265 |
+
"step": 340
|
| 266 |
+
},
|
| 267 |
+
{
|
| 268 |
+
"epoch": 2.92,
|
| 269 |
+
"grad_norm": 1.2643784284591675,
|
| 270 |
+
"learning_rate": 8.333333333333334e-06,
|
| 271 |
+
"loss": 0.3581,
|
| 272 |
+
"step": 350
|
| 273 |
+
},
|
| 274 |
+
{
|
| 275 |
+
"epoch": 3.0,
|
| 276 |
+
"grad_norm": 1.2496050596237183,
|
| 277 |
+
"learning_rate": 8.000000000000001e-06,
|
| 278 |
+
"loss": 0.3596,
|
| 279 |
+
"step": 360
|
| 280 |
+
},
|
| 281 |
+
{
|
| 282 |
+
"epoch": 3.0,
|
| 283 |
+
"eval_accuracy": 0.9538513742789277,
|
| 284 |
+
"eval_loss": 0.345228910446167,
|
| 285 |
+
"eval_runtime": 30.154,
|
| 286 |
+
"eval_samples_per_second": 97.732,
|
| 287 |
+
"eval_steps_per_second": 0.73,
|
| 288 |
+
"step": 360
|
| 289 |
+
},
|
| 290 |
+
{
|
| 291 |
+
"epoch": 3.08,
|
| 292 |
+
"grad_norm": 1.3197300434112549,
|
| 293 |
+
"learning_rate": 7.666666666666667e-06,
|
| 294 |
+
"loss": 0.3436,
|
| 295 |
+
"step": 370
|
| 296 |
+
},
|
| 297 |
+
{
|
| 298 |
+
"epoch": 3.17,
|
| 299 |
+
"grad_norm": 1.0845314264297485,
|
| 300 |
+
"learning_rate": 7.333333333333333e-06,
|
| 301 |
+
"loss": 0.3334,
|
| 302 |
+
"step": 380
|
| 303 |
+
},
|
| 304 |
+
{
|
| 305 |
+
"epoch": 3.25,
|
| 306 |
+
"grad_norm": 1.349538803100586,
|
| 307 |
+
"learning_rate": 7e-06,
|
| 308 |
+
"loss": 0.3274,
|
| 309 |
+
"step": 390
|
| 310 |
+
},
|
| 311 |
+
{
|
| 312 |
+
"epoch": 3.33,
|
| 313 |
+
"grad_norm": 1.1228528022766113,
|
| 314 |
+
"learning_rate": 6.666666666666667e-06,
|
| 315 |
+
"loss": 0.3381,
|
| 316 |
+
"step": 400
|
| 317 |
+
},
|
| 318 |
+
{
|
| 319 |
+
"epoch": 3.42,
|
| 320 |
+
"grad_norm": 0.8805223703384399,
|
| 321 |
+
"learning_rate": 6.333333333333333e-06,
|
| 322 |
+
"loss": 0.3157,
|
| 323 |
+
"step": 410
|
| 324 |
+
},
|
| 325 |
+
{
|
| 326 |
+
"epoch": 3.5,
|
| 327 |
+
"grad_norm": 1.097921371459961,
|
| 328 |
+
"learning_rate": 6e-06,
|
| 329 |
+
"loss": 0.2886,
|
| 330 |
+
"step": 420
|
| 331 |
+
},
|
| 332 |
+
{
|
| 333 |
+
"epoch": 3.58,
|
| 334 |
+
"grad_norm": 1.1974855661392212,
|
| 335 |
+
"learning_rate": 5.666666666666667e-06,
|
| 336 |
+
"loss": 0.3184,
|
| 337 |
+
"step": 430
|
| 338 |
+
},
|
| 339 |
+
{
|
| 340 |
+
"epoch": 3.67,
|
| 341 |
+
"grad_norm": 0.9140454530715942,
|
| 342 |
+
"learning_rate": 5.333333333333334e-06,
|
| 343 |
+
"loss": 0.3114,
|
| 344 |
+
"step": 440
|
| 345 |
+
},
|
| 346 |
+
{
|
| 347 |
+
"epoch": 3.75,
|
| 348 |
+
"grad_norm": 0.7140288949012756,
|
| 349 |
+
"learning_rate": 5e-06,
|
| 350 |
+
"loss": 0.2978,
|
| 351 |
+
"step": 450
|
| 352 |
+
},
|
| 353 |
+
{
|
| 354 |
+
"epoch": 3.83,
|
| 355 |
+
"grad_norm": 0.7845974564552307,
|
| 356 |
+
"learning_rate": 4.666666666666667e-06,
|
| 357 |
+
"loss": 0.3082,
|
| 358 |
+
"step": 460
|
| 359 |
+
},
|
| 360 |
+
{
|
| 361 |
+
"epoch": 3.92,
|
| 362 |
+
"grad_norm": 1.4480149745941162,
|
| 363 |
+
"learning_rate": 4.333333333333334e-06,
|
| 364 |
+
"loss": 0.3349,
|
| 365 |
+
"step": 470
|
| 366 |
+
},
|
| 367 |
+
{
|
| 368 |
+
"epoch": 4.0,
|
| 369 |
+
"grad_norm": 0.9842857718467712,
|
| 370 |
+
"learning_rate": 4.000000000000001e-06,
|
| 371 |
+
"loss": 0.299,
|
| 372 |
+
"step": 480
|
| 373 |
+
},
|
| 374 |
+
{
|
| 375 |
+
"epoch": 4.0,
|
| 376 |
+
"eval_accuracy": 0.9582626399728538,
|
| 377 |
+
"eval_loss": 0.3117567300796509,
|
| 378 |
+
"eval_runtime": 30.981,
|
| 379 |
+
"eval_samples_per_second": 95.123,
|
| 380 |
+
"eval_steps_per_second": 0.71,
|
| 381 |
+
"step": 480
|
| 382 |
+
},
|
| 383 |
+
{
|
| 384 |
+
"epoch": 4.08,
|
| 385 |
+
"grad_norm": 1.228848934173584,
|
| 386 |
+
"learning_rate": 3.6666666666666666e-06,
|
| 387 |
+
"loss": 0.3423,
|
| 388 |
+
"step": 490
|
| 389 |
+
},
|
| 390 |
+
{
|
| 391 |
+
"epoch": 4.17,
|
| 392 |
+
"grad_norm": 0.9397204518318176,
|
| 393 |
+
"learning_rate": 3.3333333333333333e-06,
|
| 394 |
+
"loss": 0.3014,
|
| 395 |
+
"step": 500
|
| 396 |
+
},
|
| 397 |
+
{
|
| 398 |
+
"epoch": 4.25,
|
| 399 |
+
"grad_norm": 1.3466633558273315,
|
| 400 |
+
"learning_rate": 3e-06,
|
| 401 |
+
"loss": 0.29,
|
| 402 |
+
"step": 510
|
| 403 |
+
},
|
| 404 |
+
{
|
| 405 |
+
"epoch": 4.33,
|
| 406 |
+
"grad_norm": 1.0849584341049194,
|
| 407 |
+
"learning_rate": 2.666666666666667e-06,
|
| 408 |
+
"loss": 0.3092,
|
| 409 |
+
"step": 520
|
| 410 |
+
},
|
| 411 |
+
{
|
| 412 |
+
"epoch": 4.42,
|
| 413 |
+
"grad_norm": 0.85384202003479,
|
| 414 |
+
"learning_rate": 2.3333333333333336e-06,
|
| 415 |
+
"loss": 0.2996,
|
| 416 |
+
"step": 530
|
| 417 |
+
},
|
| 418 |
+
{
|
| 419 |
+
"epoch": 4.5,
|
| 420 |
+
"grad_norm": 1.3900597095489502,
|
| 421 |
+
"learning_rate": 2.0000000000000003e-06,
|
| 422 |
+
"loss": 0.2788,
|
| 423 |
+
"step": 540
|
| 424 |
+
},
|
| 425 |
+
{
|
| 426 |
+
"epoch": 4.58,
|
| 427 |
+
"grad_norm": 1.5993285179138184,
|
| 428 |
+
"learning_rate": 1.6666666666666667e-06,
|
| 429 |
+
"loss": 0.3014,
|
| 430 |
+
"step": 550
|
| 431 |
+
},
|
| 432 |
+
{
|
| 433 |
+
"epoch": 4.67,
|
| 434 |
+
"grad_norm": 1.0854403972625732,
|
| 435 |
+
"learning_rate": 1.3333333333333334e-06,
|
| 436 |
+
"loss": 0.2894,
|
| 437 |
+
"step": 560
|
| 438 |
+
},
|
| 439 |
+
{
|
| 440 |
+
"epoch": 4.75,
|
| 441 |
+
"grad_norm": 1.186776041984558,
|
| 442 |
+
"learning_rate": 1.0000000000000002e-06,
|
| 443 |
+
"loss": 0.2906,
|
| 444 |
+
"step": 570
|
| 445 |
+
},
|
| 446 |
+
{
|
| 447 |
+
"epoch": 4.83,
|
| 448 |
+
"grad_norm": 1.3972039222717285,
|
| 449 |
+
"learning_rate": 6.666666666666667e-07,
|
| 450 |
+
"loss": 0.2906,
|
| 451 |
+
"step": 580
|
| 452 |
+
},
|
| 453 |
+
{
|
| 454 |
+
"epoch": 4.92,
|
| 455 |
+
"grad_norm": 1.1296278238296509,
|
| 456 |
+
"learning_rate": 3.3333333333333335e-07,
|
| 457 |
+
"loss": 0.2923,
|
| 458 |
+
"step": 590
|
| 459 |
+
},
|
| 460 |
+
{
|
| 461 |
+
"epoch": 5.0,
|
| 462 |
+
"grad_norm": 1.1308748722076416,
|
| 463 |
+
"learning_rate": 0.0,
|
| 464 |
+
"loss": 0.2729,
|
| 465 |
+
"step": 600
|
| 466 |
+
},
|
| 467 |
+
{
|
| 468 |
+
"epoch": 5.0,
|
| 469 |
+
"eval_accuracy": 0.9596199524940617,
|
| 470 |
+
"eval_loss": 0.2971116304397583,
|
| 471 |
+
"eval_runtime": 30.8812,
|
| 472 |
+
"eval_samples_per_second": 95.43,
|
| 473 |
+
"eval_steps_per_second": 0.712,
|
| 474 |
+
"step": 600
|
| 475 |
+
}
|
| 476 |
+
],
|
| 477 |
+
"logging_steps": 10,
|
| 478 |
+
"max_steps": 600,
|
| 479 |
+
"num_input_tokens_seen": 0,
|
| 480 |
+
"num_train_epochs": 5,
|
| 481 |
+
"save_steps": 500,
|
| 482 |
+
"total_flos": 6.470773575863869e+18,
|
| 483 |
+
"train_batch_size": 140,
|
| 484 |
+
"trial_name": null,
|
| 485 |
+
"trial_params": null
|
| 486 |
+
}
|
checkpoint-600/training_args.bin
ADDED
|
@@ -0,0 +1,3 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
version https://git-lfs.github.com/spec/v1
|
| 2 |
+
oid sha256:86e8111f193faa4bcd6f109cb2c94b0399e1aba4c57781d7b1aba4657f4ea302
|
| 3 |
+
size 4856
|
requirements.txt
ADDED
|
@@ -0,0 +1,4 @@
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 1 |
+
torchaudio
|
| 2 |
+
omegaconf
|
| 3 |
+
gradio
|
| 4 |
+
git+https://github.com/huggingface/transformers
|