Final test data cache - inside CI docker images (#40689)

* run

* build

* build

* fix

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>

Author: ydshieh

.circleci/create_circleci_config.py

@@ -177,7 +177,9 @@ def to_dict(self):
                     "command": f"TESTS=$(circleci tests split  --split-by=timings {self.job_name}_test_list.txt) && echo $TESTS > splitted_tests.txt && echo $TESTS | tr ' ' '\n'" if self.parallelism else f"awk '{{printf \"%s \", $0}}' {self.job_name}_test_list.txt > splitted_tests.txt"
                     }
             },
-            {"run": {"name": "fetch hub objects before pytest", "command": "python3 utils/fetch_hub_objects_for_ci.py"}},
+            # During the CircleCI docker images build time, we might already (or not) download the data.
+            # If it's done already, the files are inside the directory `/test_data/`.
+            {"run": {"name": "fetch hub objects before pytest", "command": "cp -r /test_data/* . 2>/dev/null || true; python3 utils/fetch_hub_objects_for_ci.py"}},
             {"run": {
                 "name": "Run tests",
                 "command": f"({timeout_cmd} python3 -m pytest {marker_cmd} -n {self.pytest_num_workers} {junit_flags} {repeat_on_failure_flags} {' '.join(pytest_flags)} $(cat splitted_tests.txt) | tee tests_output.txt)"}

docker/custom-tokenizers.dockerfile

@@ -2,7 +2,7 @@ FROM python:3.9-slim
 ENV PYTHONDONTWRITEBYTECODE=1
 ARG REF=main
 USER root
-RUN apt-get update && apt-get install -y libsndfile1-dev espeak-ng time git cmake wget xz-utils build-essential g++5 libprotobuf-dev protobuf-compiler git-lfs
+RUN apt-get update && apt-get install -y libsndfile1-dev espeak-ng time git cmake wget xz-utils build-essential g++5 libprotobuf-dev protobuf-compiler git-lfs curl
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv pip install --no-cache-dir -U pip setuptools
 
@@ -15,12 +15,20 @@ RUN mv catch.hpp ../libs/
 RUN cmake .. -DCMAKE_INSTALL_PREFIX=/usr/local
 RUN make install -j 10
 
+WORKDIR /
 
 RUN uv pip install --no-cache --upgrade 'torch' --index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-cache-dir  --no-deps accelerate --extra-index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install  --no-cache-dir "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[ja,testing,sentencepiece,spacy,ftfy,rjieba]" unidic unidic-lite
 # spacy is not used so not tested. Causes to failures. TODO fix later
 RUN uv run python -m unidic download
+
+# fetch test data and hub objects within CircleCI docker images to reduce even more connections
+# we don't need a full clone of `transformers` to run `fetch_hub_objects_for_ci.py`
+# the data are downloaded to the directory `/test_data` and during CircleCI's CI runtime, we need to move them to the root of `transformers`
+RUN mkdir test_data && cd test_data && curl -O https://raw.githubusercontent.com/huggingface/transformers/${REF}/utils/fetch_hub_objects_for_ci.py && python3 fetch_hub_objects_for_ci.py
+
+
 RUN uv pip uninstall transformers
 
 RUN apt-get clean && rm -rf /var/lib/apt/lists/*

docker/examples-torch.dockerfile

@@ -2,11 +2,18 @@ FROM python:3.9-slim
 ENV PYTHONDONTWRITEBYTECODE=1
 ARG REF=main
 USER root
-RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-dev espeak-ng time git g++ cmake pkg-config openssh-client git-lfs ffmpeg
+RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-dev espeak-ng time git g++ cmake pkg-config openssh-client git-lfs ffmpeg curl
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv pip install --no-cache-dir -U pip setuptools
 RUN uv pip install --no-cache-dir 'torch' 'torchaudio' 'torchvision' 'torchcodec' --index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-deps timm accelerate --extra-index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-cache-dir librosa "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[sklearn,sentencepiece,vision,testing]" seqeval albumentations jiwer
+
+# fetch test data and hub objects within CircleCI docker images to reduce even more connections
+# we don't need a full clone of `transformers` to run `fetch_hub_objects_for_ci.py`
+# the data are downloaded to the directory `/test_data` and during CircleCI's CI runtime, we need to move them to the root of `transformers`
+RUN mkdir test_data && cd test_data && curl -O https://raw.githubusercontent.com/huggingface/transformers/${REF}/utils/fetch_hub_objects_for_ci.py && python3 fetch_hub_objects_for_ci.py
+
+
 RUN uv pip uninstall transformers
 RUN apt-get clean && rm -rf /var/lib/apt/lists/*

docker/exotic-models.dockerfile

@@ -2,7 +2,7 @@ FROM python:3.9-slim
 ENV PYTHONDONTWRITEBYTECODE=1
 ARG REF=main
 USER root
-RUN apt-get update && apt-get install -y libsndfile1-dev espeak-ng time git libgl1 g++ tesseract-ocr git-lfs
+RUN apt-get update && apt-get install -y libsndfile1-dev espeak-ng time git libgl1 g++ tesseract-ocr git-lfs curl
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv pip install --no-cache-dir -U pip setuptools
 RUN uv pip install --no-cache-dir 'torch' 'torchaudio' 'torchvision' --index-url https://download.pytorch.org/whl/cpu
@@ -13,5 +13,12 @@ RUN uv pip install  --no-cache-dir "git+https://github.com/huggingface/transform
 # RUN git clone https://github.com/facebookresearch/detectron2.git
 # RUN python3 -m pip install --no-cache-dir -e detectron2
 RUN uv pip install 'git+https://github.com/facebookresearch/detectron2.git@92ae9f0b92aba5867824b4f12aa06a22a60a45d3' --no-build-isolation
+
+# fetch test data and hub objects within CircleCI docker images to reduce even more connections
+# we don't need a full clone of `transformers` to run `fetch_hub_objects_for_ci.py`
+# the data are downloaded to the directory `/test_data` and during CircleCI's CI runtime, we need to move them to the root of `transformers`
+RUN mkdir test_data && cd test_data && curl -O https://raw.githubusercontent.com/huggingface/transformers/${REF}/utils/fetch_hub_objects_for_ci.py && python3 fetch_hub_objects_for_ci.py
+
+
 RUN uv pip uninstall transformers
 RUN apt-get clean && rm -rf /var/lib/apt/lists/*

docker/pipeline-torch.dockerfile

@@ -2,10 +2,17 @@ FROM python:3.9-slim
 ENV PYTHONDONTWRITEBYTECODE=1
 ARG REF=main
 USER root
-RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-dev espeak-ng time git pkg-config openssh-client git ffmpeg
+RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-dev espeak-ng time git pkg-config openssh-client git ffmpeg curl
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv pip install --no-cache-dir -U pip setuptools
 RUN uv pip install --no-cache-dir 'torch' 'torchaudio' 'torchvision' 'torchcodec' --index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-deps timm accelerate --extra-index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-cache-dir librosa "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[sklearn,sentencepiece,vision,testing]"
+
+# fetch test data and hub objects within CircleCI docker images to reduce even more connections
+# we don't need a full clone of `transformers` to run `fetch_hub_objects_for_ci.py`
+# the data are downloaded to the directory `/test_data` and during CircleCI's CI runtime, we need to move them to the root of `transformers`
+RUN mkdir test_data && cd test_data && curl -O https://raw.githubusercontent.com/huggingface/transformers/${REF}/utils/fetch_hub_objects_for_ci.py && python3 fetch_hub_objects_for_ci.py
+
+
 RUN uv pip uninstall transformers

docker/torch-light.dockerfile

@@ -2,10 +2,16 @@ FROM python:3.9-slim
 ENV PYTHONDONTWRITEBYTECODE=1
 ARG REF=main
 USER root
-RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-dev espeak-ng time git g++ cmake pkg-config openssh-client git-lfs ffmpeg
+RUN apt-get update &&  apt-get install -y --no-install-recommends libsndfile1-dev espeak-ng time git g++ cmake pkg-config openssh-client git-lfs ffmpeg curl
 ENV UV_PYTHON=/usr/local/bin/python
 RUN pip --no-cache-dir install uv && uv pip install --no-cache-dir -U pip setuptools
 RUN uv pip install --no-cache-dir 'torch' 'torchaudio' 'torchvision' 'torchcodec' --index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-deps timm accelerate --extra-index-url https://download.pytorch.org/whl/cpu
 RUN uv pip install --no-cache-dir librosa "git+https://github.com/huggingface/transformers.git@${REF}#egg=transformers[sklearn,sentencepiece,vision,testing,tiktoken,num2words,video]"
+
+# fetch test data and hub objects within CircleCI docker images to reduce even more connections
+# we don't need a full clone of `transformers` to run `fetch_hub_objects_for_ci.py`
+# the data are downloaded to the directory `/test_data` and during CircleCI's CI runtime, we need to move them to the root of `transformers`
+RUN mkdir test_data && cd test_data && curl -O https://raw.githubusercontent.com/huggingface/transformers/${REF}/utils/fetch_hub_objects_for_ci.py && python3 fetch_hub_objects_for_ci.py
+
 RUN uv pip uninstall transformers

tests/test_processing_common.py

@@ -1212,9 +1212,11 @@ def test_apply_chat_template_video_frame_sampling(self):
         messages[0][0]["content"][0] = {
             "type": "video",
             "url": [
-                "https://www.ilankelman.org/stopsigns/australia.jpg",
-                "https://www.ilankelman.org/stopsigns/australia.jpg",
-            ],
+                url_to_local_path(
+                    "https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/transformers/tasks/australia.jpg"
+                )
+            ]
+            * 2,
         }
         out_dict_with_video = processor.apply_chat_template(
             messages,

utils/fetch_hub_objects_for_ci.py

@@ -35,6 +35,7 @@
     "https://huggingface.co/datasets/hf-internal-testing/fixtures_videos/resolve/main/tennis.mp4",
     "https://huggingface.co/datasets/raushan-testing-hf/videos-test/resolve/main/tiny_video.mp4",
     "https://thumbs.dreamstime.com/b/golden-gate-bridge-san-francisco-purple-flowers-california-echium-candicans-36805947.jpg",
+    "https://huggingface.co/datasets/raushan-testing-hf/videos-test/resolve/main/tiny_video.mp4",
 ]
 
 
@@ -59,12 +60,11 @@ def url_to_local_path(url, return_url_if_not_found=True):
     # `fatal: could not read Username for 'https://hub-ci.huggingface.co': Success`
     # But this repo. is never used in a test decorated by `is_staging_test`.
     if not _run_staging:
-        # Used in as `tests/models/auto/test_modeling_auto.py::AutoModelTest::test_dynamic_saving_from_local_repo --> _ = Repository( ... )`
-        # TODO: Remove this and the above test when `huggingface_hub v1.0` comes (where `Repository` will be removed).
-        _ = Repository(
-            local_dir="tiny-random-custom-architecture",
-            clone_from="hf-internal-testing/tiny-random-custom-architecture",
-        )
+        if not os.path.isdir("tiny-random-custom-architecture"):
+            _ = Repository(
+                local_dir="tiny-random-custom-architecture",
+                clone_from="hf-internal-testing/tiny-random-custom-architecture",
+            )
 
         # For `tests/test_tokenization_mistral_common.py:TestMistralCommonTokenizer`, which eventually calls
         # `mistral_common.tokens.tokenizers.utils.download_tokenizer_from_hf_hub` which (probably) doesn't have the cache.