nightly ci run on daily ci runner

.github/workflows/self-nightly-past-ci-caller.yml

@@ -1,134 +1,18 @@
 name: Self-hosted runner (nightly-past-ci-caller)
 
 on:
-  schedule:
-    # 2:17 am on each Sunday and Thursday
-
-    - cron: "17 2 * * 0,4"
   push:
     branches:
-      - run_nightly_ci*
+      - check_nightly_build
-      - run_past_ci*
 
 jobs:
-  build_nightly_ci_images:
+#  build_nightly_ci_images:
-    name: Build Nightly CI Docker Images
+#    name: Build Nightly CI Docker Images
-    if: (github.event_name == 'schedule') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_nightly_ci'))
+#    uses: ./.github/workflows/build-nightly-ci-docker-images.yml
-    uses: ./.github/workflows/build-nightly-ci-docker-images.yml
+#    secrets: inherit
-    secrets: inherit
 
   run_nightly_ci:
     name: Nightly CI
-    needs: [build_nightly_ci_images]
+    # needs: [build_nightly_ci_images]
     uses: ./.github/workflows/self-nightly-scheduled.yml
     secrets: inherit
-
-  run_past_ci_pytorch_1-13:
-    name: PyTorch 1.13
-    if: (cancelled() != true) && ((github.event_name == 'schedule') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci')))
-    needs: [run_nightly_ci]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: pytorch
-      version: "1.13"
-      sha: ${{ github.sha }}
-    secrets: inherit
-
-  run_past_ci_pytorch_1-12:
-    name: PyTorch 1.12
-    if: (cancelled() != true) && ((github.event_name == 'schedule') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci')))
-    needs: [run_past_ci_pytorch_1-13]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: pytorch
-      version: "1.12"
-      sha: ${{ github.sha }}
-    secrets: inherit
-
-  run_past_ci_pytorch_1-11:
-    name: PyTorch 1.11
-    if: (cancelled() != true) && ((github.event_name == 'schedule') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci')))
-    needs: [run_past_ci_pytorch_1-12]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: pytorch
-      version: "1.11"
-      sha: ${{ github.sha }}
-    secrets: inherit
-
-  run_past_ci_tensorflow_2-11:
-    name: TensorFlow 2.11
-    if: (cancelled() != true) && ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci'))
-    needs: [run_past_ci_pytorch_1-11]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: tensorflow
-      version: "2.11"
-      sha: ${{ github.sha }}
-    secrets: inherit
-
-  run_past_ci_tensorflow_2-10:
-    name: TensorFlow 2.10
-    if: (cancelled() != true) && ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci'))
-    needs: [run_past_ci_tensorflow_2-11]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: tensorflow
-      version: "2.10"
-      sha: ${{ github.sha }}
-    secrets: inherit
-
-  run_past_ci_tensorflow_2-9:
-    name: TensorFlow 2.9
-    if: (cancelled() != true) && ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci'))
-    needs: [run_past_ci_tensorflow_2-10]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: tensorflow
-      version: "2.9"
-      sha: ${{ github.sha }}
-    secrets: inherit
-
-  run_past_ci_tensorflow_2-8:
-    name: TensorFlow 2.8
-    if: (cancelled() != true) && ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci'))
-    needs: [run_past_ci_tensorflow_2-9]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: tensorflow
-      version: "2.8"
-      sha: ${{ github.sha }}
-    secrets: inherit
-
-  run_past_ci_tensorflow_2-7:
-    name: TensorFlow 2.7
-    if: (cancelled() != true) && ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci'))
-    needs: [run_past_ci_tensorflow_2-8]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: tensorflow
-      version: "2.7"
-      sha: ${{ github.sha }}
-    secrets: inherit
-
-  run_past_ci_tensorflow_2-6:
-    name: TensorFlow 2.6
-    if: (cancelled() != true) && ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci'))
-    needs: [run_past_ci_tensorflow_2-7]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: tensorflow
-      version: "2.6"
-      sha: ${{ github.sha }}
-    secrets: inherit
-
-  run_past_ci_tensorflow_2-5:
-    name: TensorFlow 2.5
-    if: (cancelled() != true) && ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci'))
-    needs: [run_past_ci_tensorflow_2-6]
-    uses: ./.github/workflows/self-past.yml
-    with:
-      framework: tensorflow
-      version: "2.5"
-      sha: ${{ github.sha }}
-    secrets: inherit

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

@@ -28,7 +28,7 @@ jobs:
     strategy:
       matrix:
         machine_type: [single-gpu, multi-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, past-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
     container:
       image: huggingface/transformers-all-latest-torch-nightly-gpu
       options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
@@ -68,7 +68,7 @@ jobs:
       matrix:
         folders: ${{ fromJson(needs.setup.outputs.matrix) }}
         machine_type: [single-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, past-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
     container:
       image: huggingface/transformers-all-latest-torch-nightly-gpu
       options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
@@ -129,7 +129,7 @@ jobs:
       matrix:
         folders: ${{ fromJson(needs.setup.outputs.matrix) }}
         machine_type: [multi-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, past-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
     container:
       image: huggingface/transformers-all-latest-torch-nightly-gpu
       options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
@@ -189,7 +189,7 @@ jobs:
       fail-fast: false
       matrix:
         machine_type: [single-gpu, multi-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, past-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
     needs: setup
     container:
       image: huggingface/transformers-pytorch-deepspeed-nightly-gpu

docker/transformers-all-latest-gpu/Dockerfile

@@ -29,7 +29,7 @@ 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 torch torchvision torchaudio --index-url https://download.pytorch.org/whl/test/cu118
 
 RUN python3 -m pip install --no-cache-dir -U tensorflow==2.13 protobuf==3.20.3 tensorflow_text tensorflow_probability