Prepare daily CI for torch 2.0.0 (#22135)

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

.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

docker/transformers-all-latest-gpu/Dockerfile

-FROM nvidia/cuda:11.6.2-cudnn8-devel-ubuntu20.04
+FROM nvidia/cuda:11.7.1-cudnn8-devel-ubuntu20.04
 LABEL maintainer="Hugging Face"
 
 ARG DEBIAN_FRONTEND=noninteractive
@@ -9,11 +9,11 @@ SHELL ["sh", "-lc"]
 # The following `ARG` are mainly used to specify the versions explicitly & directly in this docker file, and not meant
 # to be used as arguments for docker build (so far).
 
-ARG PYTORCH='1.13.1'
+ARG PYTORCH='2.0.0'
 # (not always a valid torch version)
 ARG INTEL_TORCH_EXT='1.11.0'
 # Example: `cu102`, `cu113`, etc.
-ARG CUDA='cu116'
+ARG CUDA='cu117'
 
 RUN apt update
 RUN apt install -y git libsndfile1-dev tesseract-ocr espeak-ng python3 python3-pip ffmpeg git-lfs
@@ -24,13 +24,15 @@ ARG REF=main
 RUN git clone https://github.com/huggingface/transformers && cd transformers && git checkout $REF
 RUN python3 -m pip install --no-cache-dir -e ./transformers[dev,onnxruntime]
 
-# TODO: Handle these in a python utility script
-RUN [ ${#PYTORCH} -gt 0 -a "$PYTORCH" != "pre" ] && VERSION='torch=='$PYTORCH'.*' ||  VERSION='torch'; echo "export VERSION='$VERSION'" >> ~/.profile
-RUN echo torch=$VERSION
-# `torchvision` and `torchaudio` should be installed along with `torch`, especially for nightly build.
-# Currently, let's just use their latest releases (when `torch` is installed with a release version)
-# TODO: We might need to specify proper versions that work with a specific torch version (especially for past CI).
-RUN [ "$PYTORCH" != "pre" ] && python3 -m pip install --no-cache-dir -U $VERSION torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/$CUDA || python3 -m pip install --no-cache-dir -U --pre torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/nightly/$CUDA
+## TODO: Handle these in a python utility script
+#RUN [ ${#PYTORCH} -gt 0 -a "$PYTORCH" != "pre" ] && VERSION='torch=='$PYTORCH'.*' ||  VERSION='torch'; echo "export VERSION='$VERSION'" >> ~/.profile
+#RUN echo torch=$VERSION
+## `torchvision` and `torchaudio` should be installed along with `torch`, especially for nightly build.
+## Currently, let's just use their latest releases (when `torch` is installed with a release version)
+## TODO: We might need to specify proper versions that work with a specific torch version (especially for past CI).
+#RUN [ "$PYTORCH" != "pre" ] && python3 -m pip install --no-cache-dir -U $VERSION torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/$CUDA || python3 -m pip install --no-cache-dir -U --pre torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/nightly/$CUDA
+
+RUN python3 -m pip install --no-cache-dir -U torch torchvision torchaudio --index-url https://download.pytorch.org/whl/test/cu117
 
 RUN python3 -m pip install --no-cache-dir -U tensorflow==2.11
 RUN python3 -m pip install --no-cache-dir -U tensorflow_probability
@@ -54,8 +56,8 @@ RUN python3 -m pip install --no-cache-dir bitsandbytes
 # For video model testing
 RUN python3 -m pip install --no-cache-dir decord av==9.2.0
 
-# For `dinat` model
-RUN python3 -m pip install --no-cache-dir natten -f https://shi-labs.com/natten/wheels/$CUDA/
+## For `dinat` model
+#RUN python3 -m pip install --no-cache-dir natten -f https://shi-labs.com/natten/wheels/$CUDA/
 
 # When installing in editable mode, `transformers` is not recognized as a package.
 # this line must be added in order for python to be aware of transformers.

docker/transformers-pytorch-deepspeed-latest-gpu/Dockerfile

-# https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes/rel_22-04.html#rel_22-04
-FROM nvcr.io/nvidia/pytorch:22.04-py3
+# https://docs.nvidia.com/deeplearning/frameworks/pytorch-release-notes/rel_22-08.html#rel_22-08
+FROM nvcr.io/nvidia/pytorch:22.08-py3
 LABEL maintainer="Hugging Face"
 
 ARG DEBIAN_FRONTEND=noninteractive
 
 ARG PYTORCH='1.13.1'
 # Example: `cu102`, `cu113`, etc.
-ARG CUDA='cu116'
+ARG CUDA='cu117'
 
 RUN apt -y update
 RUN apt install -y libaio-dev
@@ -18,11 +18,15 @@ RUN git clone https://github.com/huggingface/transformers && cd transformers &&
 # Install latest release PyTorch
 # (PyTorch must be installed before pre-compiling any DeepSpeed c++/cuda ops.)
 # (https://www.deepspeed.ai/tutorials/advanced-install/#pre-install-deepspeed-ops)
-RUN python3 -m pip install --no-cache-dir -U torch==$PYTORCH torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/$CUDA
+#RUN python3 -m pip install --no-cache-dir -U torch==$PYTORCH torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/$CUDA
+RUN python3 -m pip install --no-cache-dir -U torch torchvision torchaudio --index-url https://download.pytorch.org/whl/test/cu117
+
 
 RUN python3 -m pip install --no-cache-dir ./transformers[deepspeed-testing]
 
-RUN python3 -m pip install torch-tensorrt==1.3.0 --find-links https://github.com/pytorch/TensorRT/releases/expanded_assets/v1.3.0
+# This will uninstall torch 2.0.0
+# TODO: uncomment the following line once `torch-tensorrt` is ready for `torch 2.0.0`
+# RUN python3 -m pip install torch-tensorrt==1.3.0 --find-links https://github.com/pytorch/TensorRT/releases/expanded_assets/v1.3.0
 
 # recompile apex
 RUN python3 -m pip uninstall -y apex

docker/transformers-pytorch-gpu/Dockerfile

-FROM nvidia/cuda:11.6.2-cudnn8-devel-ubuntu20.04
+FROM nvidia/cuda:11.7.1-cudnn8-devel-ubuntu20.04
 LABEL maintainer="Hugging Face"
 
 ARG DEBIAN_FRONTEND=noninteractive
@@ -12,15 +12,17 @@ RUN git clone https://github.com/huggingface/transformers && cd transformers &&
 RUN python3 -m pip install --no-cache-dir -e ./transformers[dev-torch,testing,video]
 
 # If set to nothing, will install the latest version
-ARG PYTORCH='1.13.1'
+ARG PYTORCH='2.0.0'
 ARG TORCH_VISION=''
 ARG TORCH_AUDIO=''
 # Example: `cu102`, `cu113`, etc.
-ARG CUDA='cu116'
+ARG CUDA='cu117'
 
-RUN [ ${#PYTORCH} -gt 0 ] && VERSION='torch=='$PYTORCH'.*' ||  VERSION='torch'; python3 -m pip install --no-cache-dir -U $VERSION --extra-index-url https://download.pytorch.org/whl/$CUDA
-RUN [ ${#TORCH_VISION} -gt 0 ] && VERSION='torchvision=='TORCH_VISION'.*' ||  VERSION='torchvision'; python3 -m pip install --no-cache-dir -U $VERSION --extra-index-url https://download.pytorch.org/whl/$CUDA
-RUN [ ${#TORCH_AUDIO} -gt 0 ] && VERSION='torchaudio=='TORCH_AUDIO'.*' ||  VERSION='torchaudio'; python3 -m pip install --no-cache-dir -U $VERSION --extra-index-url https://download.pytorch.org/whl/$CUDA
+#RUN [ ${#PYTORCH} -gt 0 ] && VERSION='torch=='$PYTORCH'.*' ||  VERSION='torch'; python3 -m pip install --no-cache-dir -U $VERSION --extra-index-url https://download.pytorch.org/whl/$CUDA
+#RUN [ ${#TORCH_VISION} -gt 0 ] && VERSION='torchvision=='TORCH_VISION'.*' ||  VERSION='torchvision'; python3 -m pip install --no-cache-dir -U $VERSION --extra-index-url https://download.pytorch.org/whl/$CUDA
+#RUN [ ${#TORCH_AUDIO} -gt 0 ] && VERSION='torchaudio=='TORCH_AUDIO'.*' ||  VERSION='torchaudio'; python3 -m pip install --no-cache-dir -U $VERSION --extra-index-url https://download.pytorch.org/whl/$CUDA
+
+RUN python3 -m pip install --no-cache-dir -U torch torchvision torchaudio --index-url https://download.pytorch.org/whl/test/cu117
 
 RUN python3 -m pip uninstall -y tensorflow flax