build

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

@@ -0,0 +1,29 @@
+name: Self-hosted runner (nightly-ci)
+
+
+on:
+  repository_dispatch:
+  schedule:
+    - cron: "17 2 * * *"
+  push:
+    branches:
+      - build-cleanup-docker-build
+
+jobs:
+  model-ci:
+    name: Model CI
+    uses: ./.github/workflows/self-scheduled.yml
+    with:
+      job: run_models_gpu
+      slack_report_channel: "#transformers-ci-past-future"
+      runner: past-ci
+    secrets: inherit
+
+  deepspeed-ci:
+    name: DeepSpeed CI
+    uses: ./.github/workflows/self-scheduled.yml
+    with:
+      job: run_torch_cuda_extensions_gpu
+      slack_report_channel: "#transformers-ci-past-future"
+      runner: past-ci
+    secrets: inherit

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

@@ -9,7 +9,7 @@ on:
     branches:
       - run_nightly_ci*
       - run_past_ci*
-      - build-cleanup-docker-build
+      - build-cleanup-docker-build-temp
 
 jobs:
 #  build_nightly_ci_images:

.github/workflows/self-scheduled.yml

@@ -15,6 +15,9 @@ on:
       slack_report_channel:
         required: true
         type: string
+      runner:
+        required: true
+        type: string
 
 env:
   HF_HOME: /mnt/cache
@@ -38,7 +41,7 @@ jobs:
     strategy:
       matrix:
         machine_type: [single-gpu, multi-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, '${{ inputs.runner }}']
     container:
       image: huggingface/transformers-all-latest-gpu
       options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
@@ -105,7 +108,7 @@ jobs:
       fail-fast: false
       matrix:
         machine_type: [single-gpu, multi-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, '${{ inputs.runner }}']
     container:
       image: huggingface/transformers-pytorch-gpu
       options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
@@ -155,7 +158,7 @@ jobs:
       fail-fast: false
       matrix:
         machine_type: [single-gpu, multi-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, '${{ inputs.runner }}']
     container:
       image: huggingface/transformers-tensorflow-gpu
       options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
@@ -206,7 +209,7 @@ jobs:
       fail-fast: false
       matrix:
         machine_type: [single-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, '${{ inputs.runner }}']
     container:
       image: huggingface/transformers-all-latest-gpu
       options: --gpus 0 --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
@@ -257,7 +260,7 @@ jobs:
       fail-fast: false
       matrix:
         machine_type: [single-gpu, multi-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, '${{ inputs.runner }}']
     container:
       image: huggingface/transformers-pytorch-deepspeed-latest-gpu
       options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
@@ -319,7 +322,7 @@ jobs:
       matrix:
         folders: ${{ fromJson(needs.setup.outputs.quantization_matrix) }}
         machine_type: [single-gpu, multi-gpu]
-    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, daily-ci]
+    runs-on: ['${{ matrix.machine_type }}', nvidia-gpu, t4, '${{ inputs.runner }}']
     container:
       image: huggingface/transformers-quantization-latest-gpu
       options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/