Self-hosted runner with slow tests (scheduled) #450
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
name: Self-hosted runner with slow tests (scheduled) | |
on: | |
workflow_dispatch: | |
schedule: | |
- cron: "0 2 * * *" | |
env: | |
RUN_SLOW: "yes" | |
IS_GITHUB_CI: "1" | |
SLACK_API_TOKEN: ${{ secrets.SLACK_API_TOKEN }} | |
jobs: | |
run_core_tests_single_gpu: | |
runs-on: | |
group: aws-g6-4xlarge-plus | |
env: | |
CUDA_VISIBLE_DEVICES: "0" | |
TEST_TYPE: "single_gpu" | |
container: | |
image: huggingface/accelerate:gpu-nightly | |
options: --gpus all --shm-size "16gb" | |
defaults: | |
run: | |
shell: bash | |
steps: | |
- name: Update clone & pip install | |
run: | | |
source activate accelerate | |
git clone https://github.com/huggingface/accelerate; | |
cd accelerate; | |
git checkout ${{ github.sha }}; | |
pip install -e . --no-deps | |
pip install pytest-reportlog tabulate | |
- name: Show installed libraries | |
run: | | |
source activate accelerate; | |
pip freeze | |
- name: Run test on GPUs | |
working-directory: accelerate | |
run: | | |
source activate accelerate | |
make test | |
- name: Run examples on GPUs | |
working-directory: accelerate | |
if: always() | |
run: | | |
source activate accelerate | |
pip uninstall comet_ml -y | |
make test_examples | |
- name: Generate Report | |
working-directory: accelerate | |
if: always() | |
run: | | |
pip install slack_sdk tabulate | |
python utils/log_reports.py >> $GITHUB_STEP_SUMMARY | |
run_deepspeed_tests_single_gpu: | |
runs-on: | |
group: aws-g6-4xlarge-plus | |
env: | |
CUDA_VISIBLE_DEVICES: "0" | |
TEST_TYPE: "single_gpu_deepspeed" | |
container: | |
image: huggingface/accelerate:gpu-deepspeed-nightly | |
options: --gpus all --shm-size "16gb" | |
defaults: | |
run: | |
shell: bash | |
steps: | |
- name: Update clone & pip install | |
run: | | |
source activate accelerate | |
git clone https://github.com/huggingface/accelerate; | |
cd accelerate; | |
git checkout ${{ github.sha }}; | |
pip install -e . --no-deps | |
pip install pytest-reportlog tabulate | |
- name: Show installed libraries | |
run: | | |
source activate accelerate; | |
pip freeze | |
- name: Run test on GPUs | |
working-directory: accelerate | |
run: | | |
source activate accelerate | |
make test_deepspeed | |
- name: Run Integration tests on GPUs | |
working-directory: accelerate | |
if: always() | |
run: | | |
source activate accelerate | |
make test_integrations | |
- name: Run examples on GPUs | |
working-directory: accelerate | |
if: always() | |
run: | | |
source activate accelerate | |
pip uninstall comet_ml -y | |
make test_examples | |
- name: Generate Report | |
working-directory: accelerate | |
if: always() | |
run: | | |
pip install slack_sdk tabulate | |
python utils/log_reports.py >> $GITHUB_STEP_SUMMARY | |
run_core_tests_multi_gpu: | |
runs-on: | |
group: aws-g6-12xlarge-plus | |
env: | |
CUDA_VISIBLE_DEVICES: "0,1" | |
TEST_TYPE: "multi_gpu" | |
container: | |
image: huggingface/accelerate:gpu-nightly | |
options: --gpus all --shm-size "16gb" | |
defaults: | |
run: | |
shell: bash | |
steps: | |
- name: Update clone | |
run: | | |
source activate accelerate | |
git clone https://github.com/huggingface/accelerate; | |
cd accelerate; | |
git checkout ${{ github.sha }}; | |
pip install -e . --no-deps | |
pip install pytest-reportlog tabulate | |
- name: Show installed libraries | |
run: | | |
source activate accelerate; | |
pip freeze | |
- name: Run core and big modeling tests on GPUs | |
working-directory: accelerate | |
run: | | |
source activate accelerate | |
make test_core | |
make test_big_modeling | |
make test_cli | |
- name: Run Integration tests on GPUs | |
working-directory: accelerate | |
if: always() | |
run: | | |
source activate accelerate | |
make test_integrations | |
- name: Run examples on GPUs | |
working-directory: accelerate | |
if: always() | |
run: | | |
source activate accelerate | |
pip uninstall comet_ml -y | |
make test_examples | |
- name: Generate Report | |
working-directory: accelerate | |
if: always() | |
run: | | |
pip install slack_sdk tabulate | |
python utils/log_reports.py >> $GITHUB_STEP_SUMMARY | |
run_deepspeed_tests_multi_gpu: | |
runs-on: | |
group: aws-g6-12xlarge-plus | |
env: | |
CUDA_VISIBLE_DEVICES: "0,1" | |
TEST_TYPE: "multi_gpu_deepspeed" | |
container: | |
image: huggingface/accelerate:gpu-deepspeed-nightly | |
options: --gpus all --shm-size "16gb" | |
defaults: | |
run: | |
shell: bash | |
steps: | |
- name: Update clone | |
run: | | |
source activate accelerate | |
git clone https://github.com/huggingface/accelerate; | |
cd accelerate; | |
git checkout ${{ github.sha }}; | |
pip install -e . --no-deps | |
pip install pytest-reportlog tabulate | |
- name: Show installed libraries | |
run: | | |
source activate accelerate; | |
pip freeze | |
- name: Run DeepSpeed tests | |
working-directory: accelerate | |
run: | | |
source activate accelerate | |
make test_deepspeed | |
- name: Run Integration tests on GPUs | |
working-directory: accelerate | |
if: always() | |
run: | | |
source activate accelerate | |
make test_integrations | |
- name: Run examples on GPUs | |
working-directory: accelerate | |
if: always() | |
run: | | |
source activate accelerate | |
pip uninstall comet_ml -y | |
make test_examples | |
- name: Generate Report | |
working-directory: accelerate | |
if: always() | |
run: | | |
pip install slack_sdk tabulate | |
python utils/log_reports.py >> $GITHUB_STEP_SUMMARY | |
run-integration-tests: | |
if: always() | |
uses: ./.github/workflows/self_hosted_integration_tests.yml |