try

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

@@ -3,7 +3,7 @@ name: Build docker images (scheduled)
 on:
   push:
     branches:
-      - build_ci_docker_image*
+      - trigger-amd-image-build
   repository_dispatch:
   workflow_call:
     inputs:
@@ -18,186 +18,6 @@ concurrency:
   cancel-in-progress: false
 
 jobs:
-  latest-docker:
-    name: "Latest PyTorch + TensorFlow [dev]"
-    runs-on: [intel-cpu, 8-cpu, ci]
-    steps:
-      -
-        name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-      -
-        name: Check out code
-        uses: actions/checkout@v3
-      -
-        name: Login to DockerHub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_PASSWORD }}
-      -
-        name: Build and push
-        uses: docker/build-push-action@v5
-        with:
-          context: ./docker/transformers-all-latest-gpu
-          build-args: |
-            REF=main
-          push: true
-          tags: huggingface/transformers-all-latest-gpu${{ inputs.image_postfix }}
-      # Push CI images still need to be re-built daily
-      -
-        name: Build and push (for Push CI) in a daily basis
-        # This condition allows `schedule` events, or `push` events that trigger this workflow NOT via `workflow_call`.
-        # The later case is useful for manual image building for debugging purpose. Use another tag in this case!
-        if: inputs.image_postfix != '-push-ci'
-        uses: docker/build-push-action@v5
-        with:
-          context: ./docker/transformers-all-latest-gpu
-          build-args: |
-            REF=main
-          push: true
-          tags: huggingface/transformers-all-latest-gpu-push-ci
-
-  latest-torch-deepspeed-docker:
-    name: "Latest PyTorch + DeepSpeed"
-    runs-on: [intel-cpu, 8-cpu, ci]
-    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@v3
-      -
-        name: Check out code
-        uses: actions/checkout@v3
-      -
-        name: Login to DockerHub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_PASSWORD }}
-      -
-        name: Build and push
-        uses: docker/build-push-action@v5
-        with:
-          context: ./docker/transformers-pytorch-deepspeed-latest-gpu
-          build-args: |
-            REF=main
-          push: true
-          tags: huggingface/transformers-pytorch-deepspeed-latest-gpu${{ inputs.image_postfix }}
-
-  # Can't build 2 images in a single job `latest-torch-deepspeed-docker` (for `nvcr.io/nvidia`)
-  latest-torch-deepspeed-docker-for-push-ci-daily-build:
-    name: "Latest PyTorch + DeepSpeed (Push CI - Daily Build)"
-    runs-on: [intel-cpu, 8-cpu, ci]
-    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@v3
-      -
-        name: Check out code
-        uses: actions/checkout@v3
-      -
-        name: Login to DockerHub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_PASSWORD }}
-      # Push CI images still need to be re-built daily
-      -
-        name: Build and push (for Push CI) in a daily basis
-        # This condition allows `schedule` events, or `push` events that trigger this workflow NOT via `workflow_call`.
-        # The later case is useful for manual image building for debugging purpose. Use another tag in this case!
-        if: inputs.image_postfix != '-push-ci'
-        uses: docker/build-push-action@v5
-        with:
-          context: ./docker/transformers-pytorch-deepspeed-latest-gpu
-          build-args: |
-            REF=main
-          push: true
-          tags: huggingface/transformers-pytorch-deepspeed-latest-gpu-push-ci
-
-  doc-builder:
-    name: "Doc builder"
-    # Push CI doesn't need this image
-    if: inputs.image_postfix != '-push-ci'
-    runs-on: [intel-cpu, 8-cpu, ci]
-    steps:
-      -
-        name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-      -
-        name: Check out code
-        uses: actions/checkout@v3
-      -
-        name: Login to DockerHub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_PASSWORD }}
-      -
-        name: Build and push
-        uses: docker/build-push-action@v5
-        with:
-          context: ./docker/transformers-doc-builder
-          push: true
-          tags: huggingface/transformers-doc-builder
-
-  latest-pytorch:
-    name: "Latest PyTorch [dev]"
-    # Push CI doesn't need this image
-    if: inputs.image_postfix != '-push-ci'
-    runs-on: [intel-cpu, 8-cpu, ci]
-    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@v3
-      -
-        name: Check out code
-        uses: actions/checkout@v3
-      -
-        name: Login to DockerHub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_PASSWORD }}
-      -
-        name: Build and push
-        uses: docker/build-push-action@v5
-        with:
-          context: ./docker/transformers-pytorch-gpu
-          build-args: |
-            REF=main
-          push: true
-          tags: huggingface/transformers-pytorch-gpu
-
   latest-pytorch-amd:
     name: "Latest PyTorch (AMD) [dev]"
     runs-on: [intel-cpu, 8-cpu, ci]
@@ -237,34 +57,6 @@ jobs:
           push: true
           tags: huggingface/transformers-pytorch-amd-gpu-push-ci
 
-  latest-tensorflow:
-    name: "Latest TensorFlow [dev]"
-    # Push CI doesn't need this image
-    if: inputs.image_postfix != '-push-ci'
-    runs-on: [intel-cpu, 8-cpu, ci]
-    steps:
-      -
-        name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-      -
-        name: Check out code
-        uses: actions/checkout@v3
-      -
-        name: Login to DockerHub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_PASSWORD }}
-      -
-        name: Build and push
-        uses: docker/build-push-action@v5
-        with:
-          context: ./docker/transformers-tensorflow-gpu
-          build-args: |
-            REF=main
-          push: true
-          tags: huggingface/transformers-tensorflow-gpu
-
   latest-pytorch-deepspeed-amd:
     name: "PyTorch + DeepSpeed (AMD) [dev]"
     runs-on: [intel-cpu, 8-cpu, ci]
@@ -303,31 +95,3 @@ jobs:
             REF=main
           push: true
           tags: huggingface/transformers-pytorch-deepspeed-amd-gpu-push-ci
-
-  latest-quantization-torch-docker:
-    name: "Latest Pytorch + Quantization [dev]"
-     # Push CI doesn't need this image
-    if: inputs.image_postfix != '-push-ci'
-    runs-on: [intel-cpu, 8-cpu, ci]
-    steps:
-      -
-        name: Set up Docker Buildx
-        uses: docker/setup-buildx-action@v3
-      -
-        name: Check out code
-        uses: actions/checkout@v3
-      -
-        name: Login to DockerHub
-        uses: docker/login-action@v3
-        with:
-          username: ${{ secrets.DOCKERHUB_USERNAME }}
-          password: ${{ secrets.DOCKERHUB_PASSWORD }}
-      -
-        name: Build and push
-        uses: docker/build-push-action@v5
-        with:
-          context: ./docker/transformers-quantization-latest-gpu
-          build-args: |
-            REF=main
-          push: true
-          tags: huggingface/transformers-quantization-latest-gpu${{ inputs.image_postfix }}

docker/transformers-pytorch-amd-gpu/Dockerfile

@@ -34,3 +34,6 @@ RUN python3 -m pip uninstall -y tensorflow flax
 # 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.
 RUN cd transformers && python3 setup.py develop
+
+# Remove nvml as it is not compatible with ROCm
+RUN python3 -m pip uninstall py3nvml pynvml -y

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

@@ -42,4 +42,7 @@ RUN python3 -m pip install --no-cache-dir ./transformers[accelerate,testing,sent
 # this line must be added in order for python to be aware of transformers.
 RUN cd transformers && python3 setup.py develop
 
-RUN python3 -c "from deepspeed.launcher.runner import main"
+RUN python3 -c "from deepspeed.launcher.runner import main"
+
+# Remove nvml as it is not compatible with ROCm
+RUN python3 -m pip uninstall py3nvml pynvml -y