v0.41 - 2025-03-22 18:29:56 UTC - retrain-pipelines v0.1.1 - Upload model and tokenizer with README.

README.md

@@ -1,15 +1,17 @@
 ---
 # @see https://github.com/huggingface/hub-docs/blob/main/modelcard.md
 # @see https://huggingface.co/docs/huggingface_hub/guides/model-cards#update-metadata
+# @see https://huggingface.co/docs/hub/model-cards#model-card-metadata
 
-version: '0.8'
+version: '0.41'
 
-timestamp: '20250107_175611630_UTC'
+timestamp: '20250322_182956413_UTC'
 
 model_name: retrain-pipelines Function Caller
 
 base_model: unsloth/Qwen2.5-1.5B
-library_name: peft
+base_model_relation: adapter
+library_name: transformers
 datasets:
 - retrain-pipelines/func_calls
 
@@ -19,7 +21,7 @@ language:
 - en
 
 task_categories:
-- question-answering
+- text2text-generation
 
 tags:
 - retrain-pipelines
@@ -32,76 +34,147 @@ thumbnail: https://cdn-avatars.huggingface.co/v1/production/uploads/651e93137b2a
 
 
 # @see https://huggingface.co/docs/hub/models-widgets#enabling-a-widget
+# @see https://huggingface.co/docs/hub/models-widgets-examples
+# @see https://huggingface.co/docs/hub/en/model-cards#specifying-a-task--pipelinetag-
+pipeline_tag: text2text-generation
 widget:
-- text: "Is this review positive or negative? Review: Best cast iron skillet you will ever buy."
-  output:
-      text: "Hello my name is Julien"
-
-mf_run_id: '1500'
+  - text: >-
+      Hello
+    example_title: No function call
+    output:
+      text: '[]'
+  - text: >-
+      Is 49 a perfect square?
+    example_title: Perfect square
+    output:
+      text: '[{"name": "is_perfect_square", "arguments": {"num": 49}}]'
+
+mf_run_id: '1787'
 
 # @see https://huggingface.co/docs/huggingface_hub/guides/model-cards#include-evaluation-results
+# @see https://huggingface.co/docs/huggingface_hub/main/en/package_reference/cards#huggingface_hub.EvalResult
 model-index:
 - name: retrain-pipelines Function Caller
   results:
   - task:
-      type: question-answering
+      type: text2text-generation
+      name: Text2Text Generation
     dataset:
-      name: Beans
-      type: beans
+      name: retrain-pipelines Function Calling
+      type: retrain-pipelines/func_calls
+      split: validation
+      revision: 498978bb190430e136c6ab96d434c5ec2adf51f8
     metrics:
-      - type: jaccard
-        value: 0.82
-      - type: f1
-        value: 0.7633986928104575
       - type: precision
-        value: 0.73
+        value: 0.0
       - type: recall
-        value: 0.8
+        value: 0.0
+      - type: f1
+        value: 0.0
+      - type: jaccard
+        value: 0.0
 
 ---
 
-# retrain-pipelines Function Caller
-
-`version 0.8`  -  `2025-01-07 17:56:11 UTC`
-(retraining
-<a target="_blank"
-   href="https://huggingface.co/retrain-pipelines/function_caller/tree/retrain-pipelines_source-code/v0.8_20250107_175611630_UTC">source-code</a> |
-<a target="_blank"
-   href="https://huggingface.co/spaces/retrain-pipelines/online_pipeline_card_renderer/?model_repo_id=retrain-pipelines/function_caller&version_id=v0.8_20250107_175611630_UTC">pipeline-card</a>)
+<div 
+  class="
+    p-6 mb-4 rounded-lg 
+    pt-6 sm:pt-9
+    bg-gradient-to-b
+    from-purple-500 
+    dark:from-purple-500/20
+  "
+>
+  <div 
+    class="
+      pl-4 rounded-lg 
+      border-2 border-gray-100 
+      bg-gradient-to-b
+      from-purple-500 
+      dark:from-purple-500/20
+    "
+  >
+    <b>retrain-pipelines Function Caller</b>
+</div>
+  <code>version 0.41</code>  -  <code>2025-03-22 18:29:56 UTC</code>
+  (retraining
+  <a target="_blank"
+     href="https://huggingface.co/retrain-pipelines/function_caller/tree/retrain-pipelines_source-code/v0.41_20250322_182956413_UTC">source-code</a> |
+  <a target="_blank"
+     href="https://huggingface.co/spaces/retrain-pipelines/online_pipeline_card_renderer/?model_repo_id=retrain-pipelines/function_caller&version_id=v0.41_20250322_182956413_UTC">pipeline-card</a>)
+</div>
 
 Training dataset&nbsp;:
-- <code>retrain-pipelines/func_calls v0.7</code>
-(<a href="https://huggingface.co/datasets/retrain-pipelines/func_calls/blob/41d5baa3cd2377650e91e01e2a20d2a124be5e38/README.md"
-    target="_blank">41d5baa</a> -
-    2025-01-07 17:13:00 UTC)
+- <code>retrain-pipelines/func_calls v0.47</code>
+(<a href="https://huggingface.co/datasets/retrain-pipelines/func_calls/blob/498978bb190430e136c6ab96d434c5ec2adf51f8/README.md"
+    target="_blank">498978b</a> -
+    2025-03-22 18:13:14 UTC)
+    <br />
+    <img alt="" src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fhuggingface.co%2Fapi%2Fdatasets%2Fretrain-pipelines/func_calls&amp;query=%24.downloads&amp;logo=huggingface&amp;label=downloads"  class="inline-block" />&nbsp;<img alt="" src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fhuggingface.co%2Fapi%2Fdatasets%2Fretrain-pipelines/func_calls&amp;query=%24.likes&amp;logo=huggingface&amp;label=likes"  class="inline-block" />
 
 Base model&nbsp;:
 - <code>unsloth/Qwen2.5-1.5B</code>
-(<a href="https://huggingface.co/unsloth/Qwen2.5-1.5B/blob/8951671def651bbedbcdea3751f46cf35e78dfa9/README.md"
-    target="_blank">8951671</a> -
-    2024-11-12 08:07:39 UTC)
+(<a href="https://huggingface.co/unsloth/Qwen2.5-1.5B/blob/2d0a015faee2c1af360a6725a30c4d7a258ac4d4/README.md"
+    target="_blank">2d0a015</a> -
+    2025-02-06 02:32:14 UTC)
+    <br />
+    <img alt="" src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fhuggingface.co%2Fapi%2Fmodels%2Funsloth/Qwen2.5-1.5B&amp;query=%24.downloads&amp;logo=huggingface&amp;label=downloads"  class="inline-block" />&nbsp;<img alt="" src="https://img.shields.io/badge/dynamic/json?url=https%3A%2F%2Fhuggingface.co%2Fapi%2Fmodels%2Funsloth/Qwen2.5-1.5B&amp;query=%24.likes&amp;logo=huggingface&amp;label=likes"  class="inline-block" /><br />
 arxiv&nbsp;:<br />
   - <code><a href="https://huggingface.co/papers/2407.10671"
              target="_blank">2407.10671</a></code><br />
+The herein LoRa adapter can for instance be used as follows&nbsp;:<br />
+```python
+from transformers import AutoModelForCausalLM, AutoTokenizer
+from torch import device, cuda
+
+repo_id = "retrain-pipelines/function_caller"
+revision = "<model_revision_commit_hash>"
+model = AutoModelForCausalLM.from_pretrained(
+    repo_id, revision=revision, torch_dtype="auto", device_map="auto")
+tokenizer = AutoTokenizer.from_pretrained(
+    repo_id, revision=revision, torch_dtype="auto", device_map="auto")
+
+device = device("cuda" if cuda.is_available() else "cpu")
+def generate_tool_calls_list(query, max_new_tokens=400) -> str:
+    formatted_query = tokenizer.chat_template.format(query, "")
+    inputs = tokenizer(formatted_query, return_tensors="pt").input_ids.to(device)
+    outputs = model.generate(inputs, max_new_tokens=max_new_tokens, do_sample=False)
+    generated_text = tokenizer.batch_decode(outputs, skip_special_tokens=True)[0]
+    return generated_text[len(formatted_query):].strip()
+
+generate_tool_calls_list("Is 49 a perfect square ?")
+```
+
 <br />
 <br />
-<br />
-<br />
-<br />
-<br />
-&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; TODO<br />
-&nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; &nbsp; Section on Eval goes here
-<br />
-<br />
-<br />
-<br />
-<br />
-<br />
 
-<hr />
-Powered by
-<code><a target="_blank"
-         href="https://github.com/aurelienmorgan/retrain-pipelines">retrain-pipelines
-      0.1.1</a></code> - 
-<code>Run by <a target="_blank" href="https://huggingface.co/Aurelien-Morgan-Bot">Aurelien-Morgan-Bot</a></code> -
-<em><b>UnslothFuncCallFlow</b></em> - mf_run_id&nbsp;: <code>1500</code>
+<div 
+  class="
+    p-6 mb-4 rounded-lg 
+    pt-6 sm:pt-9
+    px-4
+    pb-1 
+    bg-gradient-to-t
+    from-purple-500 
+    dark:from-purple-500/20
+  "
+>
+  <div 
+    class="
+      p-6 mb-4 rounded-lg 
+      border-2 border-gray-100 
+      pt-6 sm:pt-9
+      bg-gradient-to-t
+      from-purple-500 
+      dark:from-purple-500/20
+    "
+  >
+    Powered by
+    <code><a target="_blank"
+             href="https://github.com/aurelienmorgan/retrain-pipelines">retrain-pipelines
+          0.1.1</a></code> - 
+    <code>Run by <a target="_blank" href="https://huggingface.co/Aurelien-Morgan-Bot">Aurelien-Morgan-Bot</a></code> -
+    <em><b>UnslothFuncCallFlow</b></em> - mf_run_id&nbsp;: <code>1787</code>
+  </div>
+</div>
+

adapter_config.json

@@ -23,13 +23,13 @@
   "rank_pattern": {},
   "revision": null,
   "target_modules": [
+    "up_proj",
     "q_proj",
-    "k_proj",
-    "o_proj",
     "gate_proj",
-    "up_proj",
+    "o_proj",
     "v_proj",
-    "down_proj"
+    "down_proj",
+    "k_proj"
   ],
   "task_type": "CAUSAL_LM",
   "use_dora": false,

adapter_model.safetensors

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:6247fdb2154f3690911ce04a35ed521a51c7abadae9662227916a55fc42c7d39
+oid sha256:ee7faa4ffe2eeb438b075e4338dfaf364aad215ab0a54bd74b057f3c29de4e19
 size 2457916480

added_tokens.json

@@ -1,7 +1,6 @@
 {
   "</tool_call>": 151658,
   "<tool_call>": 151657,
-  "<|PAD_TOKEN|>": 151665,
   "<|box_end|>": 151649,
   "<|box_start|>": 151648,
   "<|endoftext|>": 151643,

runs/cpt/{events.out.tfevents.1736270700.MSI.22100.0 → events.out.tfevents.1742667255.MSI.9607.0}

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:8d8192feec4b8bd56d7285b07cc3cebfb6dc789d2727b1ac0c21987583a5b900
-size 8827
+oid sha256:b60b3375aab9dde64c36065827c8eab8fd462d4ed4f5acbce52d867fb96dda50
+size 6351

runs/sft/events.out.tfevents.1742667483.MSI.10041.0

@@ -0,0 +1,3 @@
+version https://git-lfs.github.com/spec/v1
+oid sha256:645658519c807daa593913554297faa77f787fcf2fb841975da0ba8fb8b18e1f
+size 6349

special_tokens_map.json

@@ -22,7 +22,7 @@
     "single_word": false
   },
   "pad_token": {
-    "content": "<|PAD_TOKEN|>",
+    "content": "<|vision_pad|>",
     "lstrip": false,
     "normalized": false,
     "rstrip": false,

tokenizer.json

@@ -1,3 +1,3 @@
 version https://git-lfs.github.com/spec/v1
-oid sha256:fab42efe8d17406525a9154b728cf9e957629a8ed7ce997770efdd71128c6a1a
-size 11422086
+oid sha256:9c5ae00e602b8860cbd784ba82a8aa14e8feecec692e7076590d014d7b7fdafa
+size 11421896

tokenizer_config.json

@@ -177,14 +177,6 @@
       "rstrip": false,
       "single_word": false,
       "special": false
-    },
-    "151665": {
-      "content": "<|PAD_TOKEN|>",
-      "lstrip": false,
-      "normalized": false,
-      "rstrip": false,
-      "single_word": false,
-      "special": true
     }
   },
   "additional_special_tokens": [
@@ -203,11 +195,13 @@
     "<|video_pad|>"
   ],
   "bos_token": null,
+  "chat_template": "You specialize in generating tool calls. Given a query, your task is to return a list of tool calls based on your knowledge of known tools.\n\nRules:\n1. You can only use tools you know. Do not create new tools under any circumstances.\n2. If a query does not match any known tool, return an empty list ([]).\n3. If information is missing to use a known tool, do not attempt to use it.\n4. Your response must always be a valid JSON array, and nothing else.\n\nBe precise and do not guess.\n\n# query:\n    {}\n# response:\n    {}",
   "clean_up_tokenization_spaces": false,
   "eos_token": "<|endoftext|>",
   "errors": "replace",
-  "model_max_length": 131072,
-  "pad_token": "<|PAD_TOKEN|>",
+  "extra_special_tokens": {},
+  "model_max_length": 32768,
+  "pad_token": "<|vision_pad|>",
   "padding_side": "left",
   "split_special_tokens": false,
   "tokenizer_class": "Qwen2Tokenizer",