SUMIN/HuggingFace_transformer
1
0
Fork 0
Code Issues Pull Requests Actions 7 Packages Projects Releases Wiki Activity

(Re-)Enable Nightly + Past CI (#22393)

Browse Source 
* Enable Nightly + Past CI

* put schedule

---------

Co-authored-by: ydshieh <ydshieh@users.noreply.github.com>
This commit is contained in:
Yih-Dar
2023-03-30 21:06:35 +02:00
committed by GitHub
parent d5de578c22
commit 0fe6c6bdca
13 changed files with 446 additions and 349 deletions
Download Patch File Download Diff File Expand all files Collapse all files

59
.github/workflows/build-docker-images.yml vendored
View File

@@ -3,7 +3,7 @@ name: Build docker images (scheduled)
on:
push:
branches:
- docker-image*
- build_ci_docker_image*
repository_dispatch:
workflow_call:
inputs:
@@ -67,35 +67,6 @@ jobs:
push: true
tags: huggingface/transformers-all-latest-gpu-push-ci
latest-with-torch-nightly-docker:
name: "Nightly PyTorch + Stable TensorFlow"
# Push CI doesn't need this image
if: inputs.image_postfix != '-push-ci'
runs-on: ubuntu-latest
steps:
-
name: Set up Docker Buildx
uses: docker/setup-buildx-action@v2
-
name: Check out code
uses: actions/checkout@v3
-
name: Login to DockerHub
uses: docker/login-action@v2
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_PASSWORD }}
-
name: Build and push
uses: docker/build-push-action@v3
with:
context: ./docker/transformers-all-latest-gpu
build-args: |
REF=main
PYTORCH=pre
push: true
tags: huggingface/transformers-all-latest-torch-nightly-gpu
latest-torch-deepspeed-docker:
name: "Latest PyTorch + DeepSpeed"
runs-on: ubuntu-latest
@@ -153,34 +124,6 @@ jobs:
push: true
tags: huggingface/transformers-pytorch-deepspeed-latest-gpu-push-ci
nightly-torch-deepspeed-docker:
name: "Nightly PyTorch + DeepSpeed"
# Push CI doesn't need this image
if: inputs.image_postfix != '-push-ci'
runs-on: ubuntu-latest
steps:
-
name: Set up Docker Buildx
uses: docker/setup-buildx-action@v2
-
name: Check out code
uses: actions/checkout@v3
-
name: Login to DockerHub
uses: docker/login-action@v2
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_PASSWORD }}
-
name: Build and push
uses: docker/build-push-action@v3
with:
context: ./docker/transformers-pytorch-deepspeed-nightly-gpu
build-args: |
REF=main
push: true
tags: huggingface/transformers-pytorch-deepspeed-nightly-gpu
doc-builder:
name: "Doc builder"
# Push CI doesn't need this image

75
.github/workflows/build-nightly-ci-docker-images.yml vendored Normal file
View File

@@ -0,0 +1,75 @@
name: Build docker images (Nightly CI)
on:
workflow_call:
push:
branches:
- build_nightly_ci_docker_image*
concurrency:
group: docker-images-builds
cancel-in-progress: false
jobs:
latest-with-torch-nightly-docker:
name: "Nightly PyTorch + Stable TensorFlow"
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
-
name: Check out code
uses: actions/checkout@v3
-
name: Login to DockerHub
uses: docker/login-action@v2
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_PASSWORD }}
-
name: Build and push
uses: docker/build-push-action@v3
with:
context: ./docker/transformers-all-latest-gpu
build-args: |
REF=main
PYTORCH=pre
push: true
tags: huggingface/transformers-all-latest-torch-nightly-gpu
nightly-torch-deepspeed-docker:
name: "Nightly PyTorch + DeepSpeed"
runs-on: ubuntu-latest
steps:
-
name: Set up Docker Buildx
uses: docker/setup-buildx-action@v2
-
name: Check out code
uses: actions/checkout@v3
-
name: Login to DockerHub
uses: docker/login-action@v2
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_PASSWORD }}
-
name: Build and push
uses: docker/build-push-action@v3
with:
context: ./docker/transformers-pytorch-deepspeed-nightly-gpu
build-args: |
REF=main
push: true
tags: huggingface/transformers-pytorch-deepspeed-nightly-gpu

63
.github/workflows/build-past-ci-docker-images.yml vendored
View File

@@ -3,7 +3,7 @@ name: Build docker images (Past CI)
on:
push:
branches:
- past-ci-docker-image*
- build_past_ci_docker_image*
concurrency:
group: docker-images-builds
@@ -15,7 +15,7 @@ jobs:
strategy:
fail-fast: false
matrix:
version: ["1.11", "1.10", "1.9", "1.8", "1.7", "1.6", "1.5", "1.4"]
version: ["1.13", "1.12", "1.11", "1.10", "1.9"]
runs-on: ubuntu-latest
steps:
-
@@ -24,6 +24,17 @@ jobs:
-
name: Check out code
uses: actions/checkout@v3
-
id: get-base-image
name: Get Base Image
env:
framework_version: ${{ matrix.version }}
run: |
echo "base_image=$(python3 -c 'import os; from utils.past_ci_versions import past_versions_testing; base_image = past_versions_testing["pytorch"][os.environ["framework_version"]]["base_image"]; print(base_image)')" >> $GITHUB_OUTPUT
-
name: Print Base Image
run: |
echo ${{ steps.get-base-image.outputs.base_image }}
-
name: Login to DockerHub
uses: docker/login-action@v2
@@ -37,6 +48,7 @@ jobs:
context: ./docker/transformers-past-gpu
build-args: |
REF=main
BASE_DOCKER_IMAGE=${{ steps.get-base-image.outputs.base_image }}
FRAMEWORK=pytorch
VERSION=${{ matrix.version }}
push: true
@@ -47,7 +59,7 @@ jobs:
strategy:
fail-fast: false
matrix:
version: ["2.8", "2.7", "2.6", "2.5"]
version: ["2.11", "2.10", "2.9", "2.8", "2.7", "2.6", "2.5"]
runs-on: ubuntu-latest
steps:
-
@@ -56,6 +68,17 @@ jobs:
-
name: Check out code
uses: actions/checkout@v3
-
id: get-base-image
name: Get Base Image
env:
framework_version: ${{ matrix.version }}
run: |
echo "base_image=$(python3 -c 'import os; from utils.past_ci_versions import past_versions_testing; base_image = past_versions_testing["tensorflow"][os.environ["framework_version"]]["base_image"]; print(base_image)')" >> $GITHUB_OUTPUT
-
name: Print Base Image
run: |
echo ${{ steps.get-base-image.outputs.base_image }}
-
name: Login to DockerHub
uses: docker/login-action@v2
@@ -69,40 +92,8 @@ jobs:
context: ./docker/transformers-past-gpu
build-args: |
REF=main
BASE_DOCKER_IMAGE=${{ steps.get-base-image.outputs.base_image }}
FRAMEWORK=tensorflow
VERSION=${{ matrix.version }}
push: true
tags: huggingface/transformers-tensorflow-past-${{ matrix.version }}-gpu
past-tensorflow-docker-2-4:
name: "Past TensorFlow Docker"
strategy:
fail-fast: false
matrix:
version: ["2.4"]
runs-on: ubuntu-latest
steps:
-
name: Set up Docker Buildx
uses: docker/setup-buildx-action@v2
-
name: Check out code
uses: actions/checkout@v3
-
name: Login to DockerHub
uses: docker/login-action@v2
with:
username: ${{ secrets.DOCKERHUB_USERNAME }}
password: ${{ secrets.DOCKERHUB_PASSWORD }}
-
name: Build and push
uses: docker/build-push-action@v3
with:
context: ./docker/transformers-past-gpu
build-args: |
REF=main
BASE_DOCKER_IMAGE=nvidia/cuda:11.0.3-cudnn8-devel-ubuntu20.04
FRAMEWORK=tensorflow
VERSION=${{ matrix.version }}
push: true
tags: huggingface/transformers-tensorflow-past-${{ matrix.version }}-gpu

143
.github/workflows/self-nightly-past-ci-caller.yml vendored Normal file
View File

@@ -0,0 +1,143 @@
name: Self-hosted runner (nightly-past-ci-caller)
on:
schedule:
# 2 am on each Sunday and Thursday
- cron: "0 2 * * 0,4"
push:
branches:
- run_nightly_ci*
- run_past_ci*
jobs:
build_nightly_ci_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
secrets: inherit
run_nightly_ci:
name: Nightly CI
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"
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"
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"
secrets: inherit
run_past_ci_pytorch_1-10:
name: PyTorch 1.10
if: (cancelled() != true) && ((github.event_name == 'schedule') || ((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: pytorch
version: "1.10"
secrets: inherit
run_past_ci_pytorch_1-9:
name: PyTorch 1.9
if: (cancelled() != true) && ((github.event_name == 'schedule') || ((github.event_name == 'push') && startsWith(github.ref_name, 'run_past_ci')))
needs: [run_past_ci_pytorch_1-10]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.9"
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_1-9]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.11"
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"
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"
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"
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"
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"
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"
secrets: inherit

34
.github/workflows/self-nightly-scheduled.yml vendored
View File

@@ -1,4 +1,4 @@
name: Self-hosted runner (nightly)
name: Self-hosted runner (nightly-ci)
# Note that each job's dependencies go into a corresponding docker file.
#
@@ -8,9 +8,7 @@ name: Self-hosted runner (nightly)
on:
repository_dispatch:
# Disable temporarily until the test suite can be run under 12 hours.
# schedule:
# - cron: "0 16 * * *"
workflow_call:
env:
HF_HOME: /mnt/cache
@@ -33,7 +31,7 @@ jobs:
fetch-depth: 2
- name: Check Runner Status
run: python utils/check_self_hosted_runner.py --target_runners single-gpu-scheduled-ci-runner-docker,multi-gpu-scheduled-ci-runner-docker --token ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
run: python utils/check_self_hosted_runner.py --target_runners single-gpu-past-ci-runner-docker,multi-gpu-past-ci-runner-docker --token ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
check_runners:
name: Check Runners
@@ -41,7 +39,7 @@ jobs:
strategy:
matrix:
machine_type: [single-gpu, multi-gpu]
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker-past-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/
@@ -56,7 +54,7 @@ jobs:
strategy:
matrix:
machine_type: [single-gpu, multi-gpu]
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker-past-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/
@@ -96,7 +94,7 @@ jobs:
matrix:
folders: ${{ fromJson(needs.setup.outputs.matrix) }}
machine_type: [single-gpu]
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker-past-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/
@@ -143,7 +141,7 @@ jobs:
if: ${{ always() }}
uses: actions/upload-artifact@v3
with:
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_nightly
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
run_tests_multi_gpu:
@@ -153,7 +151,7 @@ jobs:
matrix:
folders: ${{ fromJson(needs.setup.outputs.matrix) }}
machine_type: [multi-gpu]
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker-past-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/
@@ -200,7 +198,7 @@ jobs:
if: ${{ always() }}
uses: actions/upload-artifact@v3
with:
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_nightly
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
run_all_tests_torch_cuda_extensions_gpu:
@@ -209,7 +207,7 @@ jobs:
fail-fast: false
matrix:
machine_type: [single-gpu, multi-gpu]
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker') }}
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker-past-ci') }}
needs: setup
container:
image: huggingface/transformers-pytorch-deepspeed-nightly-gpu
@@ -258,7 +256,7 @@ jobs:
if: ${{ always() }}
uses: actions/upload-artifact@v3
with:
name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports
name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports_postfix_nightly
path: /workspace/transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
send_results:
@@ -292,7 +290,7 @@ jobs:
CI_SLACK_CHANNEL_DUMMY_TESTS: ${{ secrets.CI_SLACK_CHANNEL_DUMMY_TESTS }}
CI_SLACK_REPORT_CHANNEL_ID: ${{ secrets.CI_SLACK_CHANNEL_ID_PAST_FUTURE }}
ACCESS_REPO_INFO_TOKEN: ${{ secrets.ACCESS_REPO_INFO_TOKEN }}
CI_EVENT: nightly-build
CI_EVENT: Nightly CI
RUNNER_STATUS: ${{ needs.check_runner_status.result }}
RUNNER_ENV_STATUS: ${{ needs.check_runners.result }}
SETUP_STATUS: ${{ needs.setup.result }}
@@ -302,3 +300,11 @@ jobs:
pip install slack_sdk
pip show slack_sdk
python utils/notification_service.py "${{ needs.setup.outputs.matrix }}"
# delete-artifact
- uses: geekyeggo/delete-artifact@v2
with:
name: |
single-*
multi-*

136
.github/workflows/self-past-caller.yml vendored
View File

@@ -1,136 +0,0 @@
name: Self-hosted runner (past-ci-caller)
on:
push:
branches:
- run-past-ci*
jobs:
run_past_ci_pytorch_1-11:
name: PyTorch 1.11
if: always()
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.11"
secrets: inherit
run_past_ci_pytorch_1-10:
name: PyTorch 1.10
if: always()
needs: [run_past_ci_pytorch_1-11]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.10"
secrets: inherit
run_past_ci_pytorch_1-9:
name: PyTorch 1.9
if: always()
needs: [run_past_ci_pytorch_1-10]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.9"
secrets: inherit
run_past_ci_pytorch_1-8:
name: PyTorch 1.8
if: always()
needs: [run_past_ci_pytorch_1-9]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.8"
secrets: inherit
run_past_ci_pytorch_1-7:
name: PyTorch 1.7
if: always()
needs: [run_past_ci_pytorch_1-8]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.7"
secrets: inherit
run_past_ci_pytorch_1-6:
name: PyTorch 1.6
if: always()
needs: [run_past_ci_pytorch_1-7]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.6"
secrets: inherit
run_past_ci_pytorch_1-5:
name: PyTorch 1.5
if: always()
needs: [run_past_ci_pytorch_1-6]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.5"
secrets: inherit
run_past_ci_pytorch_1-4:
name: PyTorch 1.4
if: always()
needs: [run_past_ci_pytorch_1-5]
uses: ./.github/workflows/self-past.yml
with:
framework: pytorch
version: "1.4"
secrets: inherit
run_past_ci_tensorflow_2-8:
name: TensorFlow 2.8
if: always()
needs: [run_past_ci_pytorch_1-4]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.8"
secrets: inherit
run_past_ci_tensorflow_2-7:
name: TensorFlow 2.7
if: always()
needs: [run_past_ci_tensorflow_2-8]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.7"
secrets: inherit
run_past_ci_tensorflow_2-6:
name: TensorFlow 2.6
if: always()
needs: [run_past_ci_tensorflow_2-7]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.6"
secrets: inherit
run_past_ci_tensorflow_2-5:
name: TensorFlow 2.5
if: always()
needs: [run_past_ci_tensorflow_2-6]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.5"
secrets: inherit
run_past_ci_tensorflow_2-4:
name: TensorFlow 2.4
if: always()
needs: [run_past_ci_tensorflow_2-5]
uses: ./.github/workflows/self-past.yml
with:
framework: tensorflow
version: "2.4"
secrets: inherit

83
.github/workflows/self-past.yml vendored
View File

@@ -1,4 +1,4 @@
name: Self-hosted runner (past)
name: Self-hosted runner (past-ci)
# Note that each job's dependencies go into a corresponding docker file.
#
@@ -157,7 +157,7 @@ jobs:
if: ${{ always() }}
uses: actions/upload-artifact@v3
with:
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_${{ inputs.framework }}-${{ inputs.version }}
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
run_tests_multi_gpu:
@@ -223,14 +223,80 @@ jobs:
if: ${{ always() }}
uses: actions/upload-artifact@v3
with:
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports
name: ${{ matrix.machine_type }}_run_all_tests_gpu_${{ env.matrix_folders }}_test_reports_postfix_${{ inputs.framework }}-${{ inputs.version }}
path: /transformers/reports/${{ matrix.machine_type }}_tests_gpu_${{ matrix.folders }}
run_all_tests_torch_cuda_extensions_gpu:
name: Torch CUDA extension tests
if: inputs.framework == 'pytorch'
strategy:
fail-fast: false
matrix:
machine_type: [single-gpu, multi-gpu]
runs-on: ${{ format('{0}-{1}', matrix.machine_type, 'docker-past-ci') }}
needs: setup
container:
image: huggingface/transformers-${{ inputs.framework }}-past-${{ inputs.version }}-gpu
options: --gpus all --shm-size "16gb" --ipc host -v /mnt/cache/.cache/huggingface:/mnt/cache/
steps:
- name: Update clone
working-directory: /transformers
run: git fetch && git checkout ${{ github.sha }}
- name: Remove cached torch extensions
run: rm -rf /github/home/.cache/torch_extensions/
# To avoid unknown test failures
- name: Pre build DeepSpeed *again*
working-directory: /
run: |
python3 -m pip uninstall -y deepspeed
rm -rf DeepSpeed
git clone https://github.com/microsoft/DeepSpeed && cd DeepSpeed && rm -rf build
DS_BUILD_CPU_ADAM=1 DS_BUILD_FUSED_ADAM=1 DS_BUILD_AIO=1 DS_BUILD_UTILS=1 python3 -m pip install . --global-option="build_ext" --global-option="-j8" --no-cache -v --disable-pip-version-check
- name: NVIDIA-SMI
run: |
nvidia-smi
- name: Environment
working-directory: /transformers
run: |
python3 utils/print_env.py
- name: Show installed libraries and their versions
working-directory: /transformers
run: pip freeze
- name: Run all tests on GPU
working-directory: /transformers
run: |
python3 -m pytest -v --make-reports=${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu tests/deepspeed tests/extended
- name: Failure short reports
if: ${{ failure() }}
continue-on-error: true
run: cat /transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu/failures_short.txt
- name: Test suite reports artifacts
if: ${{ always() }}
uses: actions/upload-artifact@v3
with:
name: ${{ matrix.machine_type }}_run_tests_torch_cuda_extensions_gpu_test_reports_postfix_${{ inputs.framework }}-${{ inputs.version }}
path: /transformers/reports/${{ matrix.machine_type }}_tests_torch_cuda_extensions_gpu
send_results:
name: Send results to webhook
runs-on: ubuntu-latest
if: always()
needs: [check_runner_status, check_runners, setup, run_tests_single_gpu, run_tests_multi_gpu]
needs: [
check_runner_status,
check_runners,
setup,
run_tests_single_gpu,
run_tests_multi_gpu,
run_all_tests_torch_cuda_extensions_gpu
]
steps:
- name: Preliminary job status
shell: bash
@@ -272,4 +338,11 @@ jobs:
uses: actions/upload-artifact@v3
with:
name: test_failure_tables_${{ inputs.framework }}-${{ inputs.version }}
path: test_failure_tables
path: test_failure_tables
# delete-artifact
- uses: geekyeggo/delete-artifact@v2
with:
name: |
single-*
multi-*

3
.github/workflows/self-scheduled.yml vendored
View File

@@ -10,6 +10,9 @@ on:
repository_dispatch:
schedule:
- cron: "0 2 * * *"
push:
branches:
- run_scheduled_ci*
env:
HF_HOME: /mnt/cache