Merge pull request huggingface#7 from huggingface/main

goog

.circleci/config.yml

@@ -12,7 +12,7 @@ jobs:
     # Ensure running with CircleCI/huggingface
     check_circleci_user:
         docker:
-            - image: cimg/python:3.7.12
+            - image: cimg/python:3.8.12
         parallelism: 1
         steps:
             - run: echo $CIRCLE_PROJECT_USERNAME
@@ -26,7 +26,7 @@ jobs:
     fetch_tests:
         working_directory: ~/transformers
         docker:
-            - image: cimg/python:3.7.12
+            - image: cimg/python:3.8.12
         parallelism: 1
         steps:
             - checkout
@@ -85,7 +85,7 @@ jobs:
     fetch_all_tests:
         working_directory: ~/transformers
         docker:
-            - image: cimg/python:3.7.12
+            - image: cimg/python:3.8.12
         parallelism: 1
         steps:
             - checkout
@@ -111,7 +111,7 @@ jobs:
     check_code_quality:
         working_directory: ~/transformers
         docker:
-            - image: cimg/python:3.7.12
+            - image: cimg/python:3.8.12
         resource_class: large
         environment:
             TRANSFORMERS_IS_CI: yes
@@ -121,8 +121,8 @@ jobs:
             - checkout
             - restore_cache:
                   keys:
-                      - v0.5-code_quality-{{ checksum "setup.py" }}
-                      - v0.5-code-quality
+                      - v0.6-code_quality-{{ checksum "setup.py" }}
+                      - v0.6-code-quality
             - run: pip install --upgrade pip
             - run: pip install .[all,quality]
             - save_cache:
@@ -144,7 +144,7 @@ jobs:
     check_repository_consistency:
         working_directory: ~/transformers
         docker:
-            - image: cimg/python:3.7.12
+            - image: cimg/python:3.8.12
         resource_class: large
         environment:
             TRANSFORMERS_IS_CI: yes
@@ -154,8 +154,8 @@ jobs:
             - checkout
             - restore_cache:
                   keys:
-                      - v0.5-repository_consistency-{{ checksum "setup.py" }}
-                      - v0.5-repository_consistency
+                      - v0.6-repository_consistency-{{ checksum "setup.py" }}
+                      - v0.6-repository_consistency
             - run: pip install --upgrade pip
             - run: pip install .[all,quality]
             - save_cache:

.circleci/create_circleci_config.py

@@ -24,17 +24,24 @@
 import yaml
 
 
-COMMON_ENV_VARIABLES = {"OMP_NUM_THREADS": 1, "TRANSFORMERS_IS_CI": True, "PYTEST_TIMEOUT": 120}
+COMMON_ENV_VARIABLES = {
+    "OMP_NUM_THREADS": 1,
+    "TRANSFORMERS_IS_CI": True,
+    "PYTEST_TIMEOUT": 120,
+    "RUN_PIPELINE_TESTS": False,
+    "RUN_PT_TF_CROSS_TESTS": False,
+    "RUN_PT_FLAX_CROSS_TESTS": False,
+}
 COMMON_PYTEST_OPTIONS = {"max-worker-restart": 0, "dist": "loadfile", "s": None}
-DEFAULT_DOCKER_IMAGE = [{"image": "cimg/python:3.7.12"}]
+DEFAULT_DOCKER_IMAGE = [{"image": "cimg/python:3.8.12"}]
 
 
 @dataclass
 class CircleCIJob:
     name: str
     additional_env: Dict[str, Any] = None
     cache_name: str = None
-    cache_version: str = "0.5"
+    cache_version: str = "0.6"
     docker_image: List[Dict[str, str]] = None
     install_steps: List[str] = None
     marker: Optional[str] = None
@@ -64,10 +71,12 @@ def __post_init__(self):
             self.parallelism = 1
 
     def to_dict(self):
+        env = COMMON_ENV_VARIABLES.copy()
+        env.update(self.additional_env)
         job = {
             "working_directory": self.working_directory,
             "docker": self.docker_image,
-            "environment": {**COMMON_ENV_VARIABLES, **self.additional_env},
+            "environment": env,
         }
         if self.resource_class is not None:
             job["resource_class"] = self.resource_class
@@ -173,7 +182,7 @@ def job_name(self):
     "torch_and_tf",
     additional_env={"RUN_PT_TF_CROSS_TESTS": True},
     install_steps=[
-        "sudo apt-get -y update && sudo apt-get install -y libsndfile1-dev espeak-ng git-lfs",
+        "sudo apt-get -y update && sudo apt-get install -y libsndfile1-dev espeak-ng git-lfs cmake",
         "git lfs install",
         "pip install --upgrade pip",
         "pip install .[sklearn,tf-cpu,torch,testing,sentencepiece,torch-speech,vision]",
@@ -215,7 +224,7 @@ def job_name(self):
 tf_job = CircleCIJob(
     "tf",
     install_steps=[
-        "sudo apt-get -y update && sudo apt-get install -y libsndfile1-dev espeak-ng",
+        "sudo apt-get -y update && sudo apt-get install -y libsndfile1-dev espeak-ng cmake",
         "pip install --upgrade pip",
         "pip install .[sklearn,tf-cpu,testing,sentencepiece,tf-speech,vision]",
         "pip install tensorflow_probability",
@@ -239,25 +248,28 @@ def job_name(self):
 
 pipelines_torch_job = CircleCIJob(
     "pipelines_torch",
+    additional_env={"RUN_PIPELINE_TESTS": True},
     install_steps=[
         "sudo apt-get -y update && sudo apt-get install -y libsndfile1-dev espeak-ng",
         "pip install --upgrade pip",
         "pip install .[sklearn,torch,testing,sentencepiece,torch-speech,vision,timm,video]",
     ],
     pytest_options={"rA": None},
-    tests_to_run="tests/pipelines/"
+    marker="is_pipeline_test",
 )
 
 
 pipelines_tf_job = CircleCIJob(
     "pipelines_tf",
+    additional_env={"RUN_PIPELINE_TESTS": True},
     install_steps=[
+        "sudo apt-get -y update && sudo apt-get install -y cmake",
         "pip install --upgrade pip",
         "pip install .[sklearn,tf-cpu,testing,sentencepiece,vision]",
         "pip install tensorflow_probability",
     ],
     pytest_options={"rA": None},
-    tests_to_run="tests/pipelines/"
+    marker="is_pipeline_test",
 )
 
 
@@ -307,6 +319,7 @@ def job_name(self):
     "examples_tensorflow",
     cache_name="tensorflow_examples",
     install_steps=[
+        "sudo apt-get -y update && sudo apt-get install -y cmake",
         "pip install --upgrade pip",
         "pip install .[sklearn,tensorflow,sentencepiece,testing]",
         "pip install -r examples/tensorflow/_tests_requirements.txt",
@@ -344,6 +357,7 @@ def job_name(self):
 onnx_job = CircleCIJob(
     "onnx",
     install_steps=[
+        "sudo apt-get -y update && sudo apt-get install -y cmake",
         "pip install --upgrade pip",
         "pip install .[torch,tf,testing,sentencepiece,onnxruntime,vision,rjieba]",
     ],
@@ -379,11 +393,11 @@ def job_name(self):
     "repo_utils",
     install_steps=[
         "pip install --upgrade pip",
-        "pip install .[quality,testing]",
+        "pip install .[quality,testing,torch]",
     ],
     parallelism=None,
     pytest_num_workers=1,
-    resource_class=None,
+    resource_class="large",
     tests_to_run="tests/repo_utils",
 )
 

.github/workflows/build-docker-images.yml

@@ -22,6 +22,16 @@ jobs:
     name: "Latest PyTorch + TensorFlow [dev]"
     runs-on: ubuntu-latest
     steps:
+      - name: Cleanup disk
+        run: |
+          sudo ls -l /usr/local/lib/
+          sudo ls -l /usr/share/
+          sudo du -sh /usr/local/lib/
+          sudo du -sh /usr/share/
+          sudo rm -rf /usr/local/lib/android
+          sudo rm -rf /usr/share/dotnet
+          sudo du -sh /usr/local/lib/
+          sudo du -sh /usr/share/
       -
         name: Set up Docker Buildx
         uses: docker/setup-buildx-action@v2

.github/workflows/check_runner_status.yml

@@ -57,6 +57,7 @@ jobs:
           CI_SLACK_CHANNEL_ID_DAILY: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
           CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
           CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
+          ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
           CI_EVENT: runner status check
           RUNNER_STATUS: ${{ needs.check_runner_status.result }}
           OFFLINE_RUNNERS: ${{ needs.check_runner_status.outputs.offline_runners }}

.github/workflows/doctests.yml

@@ -34,6 +34,9 @@ jobs:
         run: |
           python3 utils/print_env.py
 
+      - name: Show installed libraries and their versions
+        run: pip freeze
+
       - name: Prepare files for doctests
         run: |
           python3 utils/prepare_for_doc_test.py src docs

.github/workflows/self-nightly-scheduled.yml

@@ -291,6 +291,7 @@ jobs:
           CI_SLACK_CHANNEL_ID_DAILY: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
           CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
           CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID_PAST_FUTURE }}
+          ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
           CI_EVENT: nightly-build
           RUNNER_STATUS: ${{ needs.check_runner_status.result }}
           RUNNER_ENV_STATUS: ${{ needs.check_runners.result }}

.github/workflows/self-past.yml

@@ -254,6 +254,7 @@ jobs:
           CI_SLACK_CHANNEL_ID_DAILY: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
           CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
           CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID_PAST_FUTURE }}
+          ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
           CI_EVENT: Past CI - ${{ inputs.framework }}-${{ inputs.version }}
           RUNNER_STATUS: ${{ needs.check_runner_status.result }}
           RUNNER_ENV_STATUS: ${{ needs.check_runners.result }}

.github/workflows/self-push.yml

@@ -568,6 +568,7 @@ jobs:
           CI_SLACK_CHANNEL_ID_DAILY: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
           CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
           CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID }}
+          ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
           CI_EVENT: push
           CI_TITLE_PUSH: ${{ github.event.head_commit.message }}
           CI_TITLE_WORKFLOW_RUN: ${{ github.event.workflow_run.head_commit.message }}

.github/workflows/self-scheduled.yml

@@ -482,6 +482,7 @@ jobs:
           CI_SLACK_CHANNEL_ID_DAILY: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
           CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
           CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID_DAILY }}
+          ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
           CI_EVENT: scheduled
           RUNNER_STATUS: ${{ needs.check_runner_status.result }}
           RUNNER_ENV_STATUS: ${{ needs.check_runners.result }}

.github/workflows/update_tiny_models.yml

@@ -0,0 +1,47 @@
+name: Self-hosted runner (push)
+
+on:
+  push:
+    branches:
+      - update_tiny_models*
+  repository_dispatch:
+  schedule:
+    - cron: "0 2 * * *"
+
+env:
+  TOKEN: ${{ secrets.SYLVAIN_HF_TOKEN }}
+
+jobs:
+  update_tiny_models:
+    name: Update tiny models
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout transformers
+        uses: actions/checkout@v3
+        with:
+          fetch-depth: 2
+
+      - name: Install
+        run: |
+            python -m pip install -U .[dev]
+            python -m pip install -U natten
+
+      - name: Update tiny models
+        run: | 
+          python utils/update_tiny_models.py
+
+      - name: Full report
+        run: cat tiny_models/reports/tiny_model_creation_report.json
+
+      - name: Failure report
+        run: cat tiny_models/reports/simple_failed_report.txt
+
+      - name: Summary report
+        run: cat tiny_models/reports/tiny_model_summary.json
+
+      - name: Test suite reports artifacts
+        if: ${{ always() }}
+        uses: actions/upload-artifact@v3
+        with:
+          name: tiny_model_creation_reports
+          path: tiny_models/reports