Merge branch 'master' into support-clip-vision-model

.github/workflows/check_code_quality.yml

@@ -1,19 +1,11 @@
-name: check_code_quality
+name: Code Quality
 
 on:
   push:
-    branches: [ main ]
-    paths:
-      - "optimum/**.py"
-      - "tests/**.py"
-      - "examples/**.py"
+    branches: [main]
 
   pull_request:
-    branches: [ main ]
-    paths:
-      - "optimum/**.py"
-      - "tests/**.py"
-      - "examples/**.py"
+    branches: [main]
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
@@ -29,25 +21,23 @@ jobs:
 
     runs-on: ${{ matrix.os }}
     steps:
-    - uses: actions/checkout@v2
-    - name: Setup Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v2
-      with:
-        python-version: ${{ matrix.python-version }}
-    - name: Create and start a virtual environment
-      run: |
-        python -m venv venv
-        source venv/bin/activate
-    - name: Install dependencies
-      run: |
-        source venv/bin/activate
-        pip install --upgrade pip
-        pip install .[quality]
-    - name: Check style with black
-      run: |
-        source venv/bin/activate
-        black --check .
-    - name: Check style with ruff
-      run: |
-        source venv/bin/activate
-        ruff .
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Setup Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
+
+      - name: Install dependencies
+        run: |
+          pip install --upgrade pip
+          pip install .[quality]
+
+      - name: Check style with black
+        run: |
+          black --check .
+
+      - name: Check style with ruff
+        run: |
+          ruff .

.github/workflows/test_gptq.yml

@@ -1,29 +1,46 @@
-name: GPTQ Quantization / Test GPU
+name: GPTQ / Python - Test
 
 on:
   workflow_dispatch:
-  schedule:
-    - cron: 0 1 */3 * * # at 1am every 3 days
+  push:
+    branches: [main]
+    paths:
+      - tests/gptq/**
+      - optimum/gptq/**
+      - .github/workflows/test_gptq.yml
   pull_request:
-    types: [opened, synchronize, reopened, labeled]
-  # uncomment to enable on PR merge on main branch:
-  #push:
-  #  branches:
-  #    - main
+    branches: [main]
+    paths:
+      - tests/gptq/**
+      - optimum/gptq/**
+      - .github/workflows/test_gptq.yml
+  schedule:
+    # every day at midnight
+    - cron: "0 0 * * *"
 
 jobs:
-  do-the-job:
-    if: ${{ (github.event_name == 'workflow_dispatch') || (github.event_name == 'schedule') || contains( github.event.pull_request.labels.*.name, 'gpu-test') }}
-    name: Start self-hosted EC2 runner
+  test_gptq:
     runs-on: [single-gpu, nvidia-gpu, t4, ci]
-    env:
-      AWS_REGION: us-east-1
+
     steps:
-      - name: Checkout
-        uses: actions/checkout@v2
-      - name: Build image
-        run: |
-          docker build -f tests/gptq/Dockerfile_quantization_gpu -t gptq-gpu .
-      - name: Test with unittest within docker container
-        run: |
-          docker run --rm --gpus all -v $(pwd)/hf_cache:/root/.cache/huggingface --workdir=/workspace/optimum/tests gptq-gpu:latest
+      - name: Checkout code
+        uses: actions/checkout@v4
+
+      - name: Run tests
+        uses: addnab/docker-run-action@v3
+        with:
+          image: pytorch/pytorch:2.2.2-cuda12.1-cudnn8-runtime
+          # latest auto-gptq was built with pytorch 2.2 and cuda 12.1
+          options: |
+            --rm
+            --gpus all
+            --shm-size 16G
+            --env RUN_SLOW=1
+            --env HF_HOME=/mnt/cache/
+            --volume /mnt/cache/:/mnt/cache/
+            --volume ${{ github.workspace }}:/workspace
+            --workdir /workspace
+          run: |
+            pip install auto-gptq
+            pip install -e .[tests]
+            pytest tests/gptq -s -vvvv --durations=0

.github/workflows/test_offline.yml

@@ -2,9 +2,9 @@ name: Offline usage / Python - Test
 
 on:
   push:
-    branches: [ main ]
+    branches: [main]
   pull_request:
-    branches: [ main ]
+    branches: [main]
 
 concurrency:
   group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
@@ -15,29 +15,33 @@ jobs:
     strategy:
       fail-fast: false
       matrix:
-        python-version: [3.9]
+        python-version: [3.8, 3.9]
         os: [ubuntu-20.04]
 
     runs-on: ${{ matrix.os }}
     steps:
-    - uses: actions/checkout@v2
-    - name: Setup Python ${{ matrix.python-version }}
-      uses: actions/setup-python@v2
-      with:
-        python-version: ${{ matrix.python-version }}
-    - name: Install dependencies for pytorch export
-      run: |
-        pip install .[tests,exporters,onnxruntime]
-    - name: Test with unittest
-      run: |
-        HF_HOME=/tmp/ huggingface-cli download hf-internal-testing/tiny-random-gpt2
+      - name: Checkout code
+        uses: actions/checkout@v4
 
-        HF_HOME=/tmp/ HF_HUB_OFFLINE=1 optimum-cli export onnx --model hf-internal-testing/tiny-random-gpt2 gpt2_onnx --task text-generation
+      - name: Setup Python ${{ matrix.python-version }}
+        uses: actions/setup-python@v5
+        with:
+          python-version: ${{ matrix.python-version }}
 
-        huggingface-cli download hf-internal-testing/tiny-random-gpt2
+      - name: Install dependencies for pytorch export
+        run: |
+          pip install .[tests,exporters,onnxruntime]
 
-        HF_HUB_OFFLINE=1 optimum-cli export onnx --model hf-internal-testing/tiny-random-gpt2 gpt2_onnx --task text-generation
+      - name: Test with pytest
+        run: |
+          HF_HOME=/tmp/ huggingface-cli download hf-internal-testing/tiny-random-gpt2
 
-        pytest tests/onnxruntime/test_modeling.py -k "test_load_model_from_hub and not from_hub_onnx" -s -vvvvv
+          HF_HOME=/tmp/ HF_HUB_OFFLINE=1 optimum-cli export onnx --model hf-internal-testing/tiny-random-gpt2 gpt2_onnx --task text-generation
 
-        HF_HUB_OFFLINE=1 pytest tests/onnxruntime/test_modeling.py -k "test_load_model_from_hub and not from_hub_onnx" -s -vvvvv
+          huggingface-cli download hf-internal-testing/tiny-random-gpt2
+
+          HF_HUB_OFFLINE=1 optimum-cli export onnx --model hf-internal-testing/tiny-random-gpt2 gpt2_onnx --task text-generation
+
+          pytest tests/onnxruntime/test_modeling.py -k "test_load_model_from_hub and not from_hub_onnx" -s -vvvvv
+
+          HF_HUB_OFFLINE=1 pytest tests/onnxruntime/test_modeling.py -k "test_load_model_from_hub and not from_hub_onnx" -s -vvvvv

docs/source/_redirects.yml

@@ -19,12 +19,18 @@ habana/tutorials/pretraining: habana/usage_guides/pretraining
 
 # Optimum Intel
 intel_index: intel/index
-intel_quickstart: intel/optimization_inc
-intel_configuration: intel/reference_inc
-intel_optimization: intel/optimization_inc
-intel_quantization: intel/optimization_inc
-intel_pruning: intel/optimization_inc
-intel_trainer: intel/reference_inc
+intel_quickstart: intel/index
+intel_configuration: intel/neural_compressor/reference
+intel_optimization: intel/neural_compressor/optimization
+intel_quantization: intel/neural_compressor/optimization
+intel_pruning: intel/neural_compressor/optimization
+intel_trainer: intel/neural_compressor/reference
+intel/inference: intel/openvino/inference
+intel/optimization_ov: intel/openvino/optimization
+intel/reference_ov: intel/openvino/reference
+intel/optimization_inc: intel/neural_compressor/optimization
+intel/distributed_training: intel/neural_compressor/distributed_training
+intel/reference_inc: intel/neural_compressor/reference
 
 # Optimum Neuron
 docs/optimum-neuron/index: /docs/optimum-neuron/index

optimum/configuration_utils.py

@@ -18,9 +18,9 @@
 import json
 import os
 import re
+import warnings
 from typing import Any, Dict, List, Tuple, Union
 
-from huggingface_hub import HfFolder
 from packaging import version
 from transformers import PretrainedConfig
 from transformers import __version__ as transformers_version_str
@@ -93,7 +93,19 @@ def save_pretrained(self, save_directory: Union[str, os.PathLike], push_to_hub:
                 repo_id = self._create_repo(repo_id, **kwargs)
 
                 use_auth_token = kwargs.get("use_auth_token", None)
-                token = HfFolder.get_token() if use_auth_token is True else use_auth_token
+                token = kwargs.get("token", None)
+
+                if use_auth_token is not None:
+                    warnings.warn(
+                        "The `use_auth_token` argument is deprecated and will be removed soon. Please use the `token` argument instead.",
+                        FutureWarning,
+                    )
+                    if token is not None:
+                        raise ValueError(
+                            "You cannot use both `use_auth_token` and `token` arguments at the same time."
+                        )
+                    kwargs["token"] = use_auth_token
+                    token = use_auth_token
 
                 files_timestamps = self._get_files_timestamps(save_directory)
 
@@ -197,6 +209,7 @@ def _get_config_dict(
         resume_download = kwargs.pop("resume_download", False)
         proxies = kwargs.pop("proxies", None)
         use_auth_token = kwargs.pop("use_auth_token", None)
+        token = kwargs.pop("token", None)
         local_files_only = kwargs.pop("local_files_only", False)
         revision = kwargs.pop("revision", None)
         trust_remote_code = kwargs.pop("trust_remote_code", None)
@@ -205,6 +218,15 @@ def _get_config_dict(
         from_auto_class = kwargs.pop("_from_auto", False)
         commit_hash = kwargs.pop("_commit_hash", None)
 
+        if use_auth_token is not None:
+            warnings.warn(
+                "The `use_auth_token` argument is deprecated and will be removed soon. Please use the `token` argument instead.",
+                FutureWarning,
+            )
+            if token is not None:
+                raise ValueError("You cannot use both `use_auth_token` and `token` arguments at the same time.")
+            token = use_auth_token
+
         if trust_remote_code is True:
             logger.warning(
                 "The argument `trust_remote_code` is to be used with Auto classes. It has no effect here and is"
@@ -255,7 +277,7 @@ def _get_config_dict(
                         proxies=proxies,
                         resume_download=resume_download,
                         local_files_only=local_files_only,
-                        use_auth_token=use_auth_token,
+                        token=token,
                         user_agent=user_agent,
                     )
                 else:
@@ -268,7 +290,7 @@ def _get_config_dict(
                         proxies=proxies,
                         resume_download=resume_download,
                         local_files_only=local_files_only,
-                        use_auth_token=use_auth_token,
+                        token=token,
                         user_agent=user_agent,
                         revision=revision,
                         subfolder=subfolder,

optimum/exporters/onnx/__main__.py

@@ -15,6 +15,7 @@
 """Entry point to the optimum.exporters.onnx command line."""
 
 import argparse
+import warnings
 from pathlib import Path
 
 from huggingface_hub.constants import HUGGINGFACE_HUB_CACHE
@@ -66,6 +67,7 @@ def main_export(
     force_download: bool = False,
     local_files_only: bool = False,
     use_auth_token: Optional[Union[bool, str]] = None,
+    token: Optional[Union[bool, str]] = None,
     for_ort: bool = False,
     do_validation: bool = True,
     model_kwargs: Optional[Dict[str, Any]] = None,
@@ -135,9 +137,11 @@ def main_export(
             cached versions if they exist.
         local_files_only (`Optional[bool]`, defaults to `False`):
             Whether or not to only look at local files (i.e., do not try to download the model).
-        use_auth_token (`Optional[str]`, defaults to `None`):
+        use_auth_token (`Optional[Union[bool,str]]`, defaults to `None`):
+            Deprecated. Please use the `token` argument instead.
+        token (`Optional[Union[bool,str]]`, defaults to `None`):
             The token to use as HTTP bearer authorization for remote files. If `True`, will use the token generated
-            when running `transformers-cli login` (stored in `~/.huggingface`).
+            when running `huggingface-cli login` (stored in `huggingface_hub.constants.HF_TOKEN_PATH`).
         model_kwargs (`Optional[Dict[str, Any]]`, defaults to `None`):
             Experimental usage: keyword arguments to pass to the model during
             the export. This argument should be used along the `custom_onnx_configs` argument
@@ -174,6 +178,15 @@ def main_export(
     ```
     """
 
+    if use_auth_token is not None:
+        warnings.warn(
+            "The `use_auth_token` argument is deprecated and will be removed soon. Please use the `token` argument instead.",
+            FutureWarning,
+        )
+        if token is not None:
+            raise ValueError("You cannot use both `use_auth_token` and `token` arguments at the same time.")
+        token = use_auth_token
+
     if fp16:
         if dtype is not None:
             raise ValueError(
@@ -250,7 +263,7 @@ def main_export(
             subfolder=subfolder,
             revision=revision,
             cache_dir=cache_dir,
-            use_auth_token=use_auth_token,
+            token=token,
             local_files_only=local_files_only,
             force_download=force_download,
             trust_remote_code=trust_remote_code,
@@ -283,7 +296,7 @@ def main_export(
         subfolder=subfolder,
         revision=revision,
         cache_dir=cache_dir,
-        use_auth_token=use_auth_token,
+        token=token,
         local_files_only=local_files_only,
         force_download=force_download,
         trust_remote_code=trust_remote_code,