Merge branch 'main' into smangrul/fix-fsdp-full-state-dict-resume

huggingface · Dec 15, 2023 · fe81718 · fe81718
2 parents 239641c + ffa04de
commit fe81718
Show file tree

Hide file tree

Showing 368 changed files with 23,942 additions and 2,506 deletions.
diff --git a/.github/workflows/build-docker-images.yml b/.github/workflows/build-docker-images.yml
@@ -271,3 +271,39 @@ jobs:
             REF=main
           push: true
           tags: huggingface/transformers-tensorflow-gpu
+
+  # latest-pytorch-deepspeed-amd:
+  #   name: "PyTorch + DeepSpeed (AMD) [dev]"
+
+  #   runs-on: [self-hosted, docker-gpu, amd-gpu, single-gpu, mi210]
+  #   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: ${{ secrets.DOCKERHUB_PASSWORD }}
+  #     - name: Build and push
+  #       uses: docker/build-push-action@v5
+  #       with:
+  #         context: ./docker/transformers-pytorch-deepspeed-amd-gpu
+  #         build-args: |
+  #           REF=main
+  #         push: true
+  #         tags: huggingface/transformers-pytorch-deepspeed-amd-gpu${{ 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-pytorch-deepspeed-amd-gpu
+  #         build-args: |
+  #           REF=main
+  #         push: true
+  #         tags: huggingface/transformers-pytorch-deepspeed-amd-gpu-push-ci
diff --git a/.github/workflows/self-nightly-scheduled.yml b/.github/workflows/self-nightly-scheduled.yml
@@ -212,7 +212,7 @@ jobs:
           python3 -m pip uninstall -y deepspeed
           rm -rf DeepSpeed
           git clone https://github.com/microsoft/DeepSpeed && cd DeepSpeed && rm -rf build
-          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 DS_BUILD_UTILS=1 python3 -m pip install . --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
+          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 python3 -m pip install . --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
 
       - name: NVIDIA-SMI
         run: |
@@ -286,4 +286,4 @@ jobs:
         with:
           name: |
               single-*
-              multi-*
+              multi-*
diff --git a/.github/workflows/self-past.yml b/.github/workflows/self-past.yml
@@ -267,7 +267,7 @@ jobs:
           python3 -m pip uninstall -y deepspeed
           rm -rf DeepSpeed
           git clone https://github.com/microsoft/DeepSpeed && cd DeepSpeed && rm -rf build
-          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 DS_BUILD_UTILS=1 python3 -m pip install . --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
+          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 python3 -m pip install . --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
 
       - name: NVIDIA-SMI
         run: |
@@ -353,4 +353,4 @@ jobs:
         with:
           name: |
               single-*
-              multi-*
+              multi-*
diff --git a/.github/workflows/self-push-amd-mi210-caller.yml b/.github/workflows/self-push-amd-mi210-caller.yml
@@ -18,7 +18,7 @@ on:
 jobs:
   run_amd_ci:
     name: AMD mi210
-    if: (cancelled() != true) && ((github.event_name == 'push') && (github.ref_name == 'main' || startsWith(github.ref_name, 'run_amd_push_ci_caller')))
+    if: (cancelled() != true) && ((github.event_name == 'workflow_run') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_amd_push_ci_caller')))
     uses: ./.github/workflows/self-push-amd.yml
     with:
       gpu_flavor: mi210

diff --git a/.github/workflows/self-push-amd-mi250-caller.yml b/.github/workflows/self-push-amd-mi250-caller.yml
@@ -18,7 +18,7 @@ on:
 jobs:
   run_amd_ci:
     name: AMD mi250
-    if: (cancelled() != true) && ((github.event_name == 'push') && (github.ref_name == 'main' || startsWith(github.ref_name, 'run_amd_push_ci_caller')))
+    if: (cancelled() != true) && ((github.event_name == 'workflow_run') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_amd_push_ci_caller')))
     uses: ./.github/workflows/self-push-amd.yml
     with:
       gpu_flavor: mi250

diff --git a/.github/workflows/self-push.yml b/.github/workflows/self-push.yml
@@ -366,7 +366,7 @@ jobs:
         working-directory: /workspace
         run: |
           python3 -m pip uninstall -y deepspeed
-          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 DS_BUILD_UTILS=1 python3 -m pip install deepspeed --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
+          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 python3 -m pip install deepspeed --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
 
       - name: NVIDIA-SMI
         run: |
@@ -456,7 +456,7 @@ jobs:
         working-directory: /workspace
         run: |
           python3 -m pip uninstall -y deepspeed
-          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 DS_BUILD_UTILS=1 python3 -m pip install deepspeed --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
+          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 python3 -m pip install deepspeed --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
 
       - name: NVIDIA-SMI
         run: |

diff --git a/.github/workflows/self-scheduled-amd-mi210-caller.yml b/.github/workflows/self-scheduled-amd-mi210-caller.yml
@@ -12,7 +12,7 @@ on:
 jobs:
   run_amd_ci:
     name: AMD mi210
-    if: (cancelled() != true) && ((github.event_name == 'schedule') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_amd_scheduled_ci_caller')))
+    if: (cancelled() != true) && ((github.event_name == 'workflow_run') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_amd_scheduled_ci_caller')))
     uses: ./.github/workflows/self-scheduled-amd.yml
     with:
       gpu_flavor: mi210

diff --git a/.github/workflows/self-scheduled-amd-mi250-caller.yml b/.github/workflows/self-scheduled-amd-mi250-caller.yml
@@ -12,7 +12,7 @@ on:
 jobs:
   run_amd_ci:
     name: AMD mi250
-    if: (cancelled() != true) && ((github.event_name == 'schedule') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_amd_scheduled_ci_caller')))
+    if: (cancelled() != true) && ((github.event_name == 'workflow_run') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_amd_scheduled_ci_caller')))
     uses: ./.github/workflows/self-scheduled-amd.yml
     with:
       gpu_flavor: mi250

diff --git a/.github/workflows/self-scheduled-amd.yml b/.github/workflows/self-scheduled-amd.yml
@@ -356,6 +356,63 @@ jobs:
           name: ${{ matrix.machine_type }}_run_tests_torch_pipeline_gpu
           path: /transformers/reports/${{ matrix.machine_type }}_tests_torch_pipeline_gpu
 
+  run_tests_torch_deepspeed_gpu:
+    name: Torch ROCm deepspeed tests
+    strategy:
+      fail-fast: false
+      matrix:
+        machine_type: [single-gpu, multi-gpu]
+
+    runs-on: [self-hosted, docker-gpu, amd-gpu, '${{ matrix.machine_type }}', '${{ inputs.gpu_flavor }}']
+    needs: setup
+    container:
+      image: huggingface/transformers-pytorch-deepspeed-amd-gpu
+      options: --device /dev/kfd --device /dev/dri --env ROCR_VISIBLE_DEVICES --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
+    steps:
+      - name: Update clone
+        working-directory: /transformers
+        run: git fetch && git checkout ${{ github.sha }}
+
+      - name: Reinstall transformers in edit mode (remove the one installed during docker image build)
+        working-directory: /transformers
+        run: python3 -m pip uninstall -y transformers && python3 -m pip install -e .
+
+      - name: ROCM-SMI
+        run: |
+          rocm-smi
+      - name: ROCM-INFO
+        run: |
+          rocminfo  | grep "Agent" -A 14
+
+      - name: Show ROCR environment
+        run: |
+          echo "ROCR: $ROCR_VISIBLE_DEVICES"
+
+      - name: Environment
+        working-directory: /transformers
+        run: |
+          python3 utils/print_env.py
+
+      - name: Show installed libraries and their versions
+        working-directory: /transformers
+        run: pip freeze
+
+      - name: Run all tests on GPU
+        working-directory: /transformers
+        run: python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_torch_deepspeed_gpu tests/deepspeed tests/extended
+
+      - name: Failure short reports
+        if: ${{ failure() }}
+        continue-on-error: true
+        run: cat /transformers/reports/${{ matrix.machine_type }}_tests_torch_deepspeed_gpu/failures_short.txt
+
+      - name: Test suite reports artifacts
+        if: ${{ always() }}
+        uses: actions/upload-artifact@v3
+        with:
+          name: ${{ matrix.machine_type }}_run_tests_torch_deepspeed_gpu_test_reports
+          path: /transformers/reports/${{ matrix.machine_type }}_tests_torch_deepspeed_gpu
+
   run_extract_warnings:
     name: Extract warnings in CI artifacts
     runs-on: ubuntu-22.04
@@ -368,7 +425,7 @@ jobs:
       run_tests_multi_gpu,
       run_examples_gpu,
       run_pipelines_torch_gpu,
-      # run_all_tests_torch_cuda_extensions_gpu
+      run_tests_torch_deepspeed_gpu
     ]
     steps:
       - name: Checkout transformers
@@ -417,7 +474,7 @@ jobs:
       run_tests_multi_gpu,
       run_examples_gpu,
       run_pipelines_torch_gpu,
-      # run_all_tests_torch_cuda_extensions_gpu,
+      run_tests_torch_deepspeed_gpu,
       run_extract_warnings
     ]
     steps:

diff --git a/.github/workflows/self-scheduled.yml b/.github/workflows/self-scheduled.yml
@@ -366,7 +366,7 @@ jobs:
         working-directory: /workspace
         run: |
           python3 -m pip uninstall -y deepspeed
-          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 DS_BUILD_UTILS=1 python3 -m pip install deepspeed --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
+          DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 python3 -m pip install deepspeed --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
 
       - name: NVIDIA-SMI
         run: |
@@ -494,5 +494,5 @@ jobs:
         if: ${{ always() }}
         uses: actions/upload-artifact@v3
         with:
-          name: test_failure_tables
-          path: test_failure_tables
+          name: prev_ci_results
+          path: prev_ci_results