Fix the FSDP.optim_state_dict_to_load OOM (#3184) #1375
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: Daily | |
| on: | |
| schedule: | |
| - cron: "30 2 * * *" # 2:30 every day | |
| push: | |
| branches: | |
| - dev | |
| - main | |
| - release/** | |
| workflow_dispatch: | |
| # Cancel old runs when a new commit is pushed to the same branch if not on main or dev | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} | |
| cancel-in-progress: ${{ github.ref != 'refs/heads/main' && github.ref != 'refs/heads/dev' }} | |
| jobs: | |
| daily-pytest-cpu: | |
| uses: mosaicml/ci-testing/.github/workflows/[email protected] | |
| strategy: | |
| matrix: | |
| include: | |
| - name: cpu-3.10-2.0 | |
| container: mosaicml/pytorch:2.0.1_cpu-python3.10-ubuntu20.04 | |
| markers: not daily and (remote or not remote) and not gpu and not doctest | |
| pytest_command: coverage run -m pytest | |
| composer_package_name: mosaicml | |
| - name: cpu-3.10-2.1 | |
| container: mosaicml/pytorch:2.1.2_cpu-python3.10-ubuntu20.04 | |
| markers: not daily and (remote or not remote) and not gpu and not doctest | |
| pytest_command: coverage run -m pytest | |
| composer_package_name: mosaicml | |
| - name: cpu-3.10-2.1-composer | |
| container: mosaicml/pytorch:2.1.2_cpu-python3.10-ubuntu20.04 | |
| markers: not daily and (remote or not remote) and not gpu and not doctest | |
| pytest_command: coverage run -m pytest | |
| composer_package_name: composer | |
| - name: cpu-3.11-2.2 | |
| container: mosaicml/pytorch:2.2.0_cpu-python3.11-ubuntu20.04 | |
| markers: not daily and (remote or not remote) and not gpu and not doctest | |
| pytest_command: coverage run -m pytest | |
| composer_package_name: mosaicml | |
| - name: cpu-doctest | |
| container: mosaicml/pytorch:2.1.2_cpu-python3.10-ubuntu20.04 | |
| markers: not daily and (remote or not remote) and not gpu and doctest | |
| pytest_command: coverage run -m pytest tests/test_docs.py | |
| composer_package_name: mosaicml | |
| - name: daily-cpu-3.10-2.0 | |
| container: mosaicml/pytorch:2.0.1_cpu-python3.10-ubuntu20.04 | |
| markers: daily and (remote or not remote) and not gpu and not doctest | |
| pytest_command: coverage run -m pytest | |
| composer_package_name: mosaicml | |
| - name: daily-cpu-3.10-2.1 | |
| container: mosaicml/pytorch:2.1.2_cpu-python3.10-ubuntu20.04 | |
| markers: daily and (remote or not remote) and not gpu and not doctest | |
| pytest_command: coverage run -m pytest | |
| composer_package_name: mosaicml | |
| - name: daily-cpu-3.10-2.1-composer | |
| container: mosaicml/pytorch:2.1.2_cpu-python3.10-ubuntu20.04 | |
| markers: daily and (remote or not remote) and not gpu and not doctest | |
| pytest_command: coverage run -m pytest | |
| composer_package_name: composer | |
| - name: daily-cpu-3.11-2.2 | |
| container: mosaicml/pytorch:2.2.0_cpu-python3.11-ubuntu20.04 | |
| markers: daily and (remote or not remote) and not gpu and not doctest | |
| pytest_command: coverage run -m pytest | |
| composer_package_name: mosaicml | |
| - name: daily-cpu-doctest | |
| container: mosaicml/pytorch:2.1.2_cpu-python3.10-ubuntu20.04 | |
| markers: daily and (remote or not remote) and not gpu and doctest | |
| pytest_command: coverage run -m pytest tests/test_docs.py | |
| composer_package_name: mosaicml | |
| name: ${{ matrix.name }} | |
| if: github.repository_owner == 'mosaicml' | |
| with: | |
| container: ${{ matrix.container }} | |
| name: ${{ matrix.name }} | |
| pip_deps: "[all]" | |
| pytest-command: ${{ matrix.pytest_command }} | |
| pytest-markers: ${{ matrix.markers }} | |
| composer_package_name: ${{ matrix.composer_package_name }} | |
| pytest-wandb-entity: "mosaicml-public-integration-tests" | |
| pytest-wandb-project: "integration-tests-${{ github.sha }}" | |
| safe_directory: composer | |
| secrets: | |
| aws-secret-access-key: ${{ secrets.AWS_SECRET_ACCESS_KEY }} | |
| aws-access-key-id: ${{ secrets.AWS_ACCESS_KEY_ID }} | |
| wandb-api-key: ${{ secrets.WANDB_API_KEY }} | |
| code-eval-device: ${{ secrets.CODE_EVAL_DEVICE }} | |
| code-eval-url: ${{ secrets.CODE_EVAL_URL }} | |
| code-eval-apikey: ${{ secrets.CODE_EVAL_APIKEY }} | |
| gcs-key: ${{ secrets.GCS_KEY }} | |
| gcs-secret: ${{ secrets.GCS_SECRET }} | |
| azure-account-name: ${{ secrets.AZURE_ACCOUNT_NAME }} | |
| azure-account-access-key: ${{ secrets.AZURE_ACCOUNT_ACCESS_KEY }} | |
| coverage: | |
| uses: ./.github/workflows/coverage.yaml | |
| name: Coverage Results | |
| if: github.repository_owner == 'mosaicml' | |
| needs: [daily-pytest-cpu] | |
| with: | |
| download-path: artifacts | |
| daily-pytest-gpu: | |
| uses: mosaicml/ci-testing/.github/workflows/[email protected] | |
| strategy: | |
| matrix: | |
| # Unlike CPU tests, we run daily tests together with GPU tests to minimize launch time | |
| # on MCLOUD and not eat up all GPUs at once | |
| include: | |
| - name: "gpu-3.10-2.0" | |
| container: mosaicml/pytorch_vision:2.0.1_cu117-python3.10-ubuntu20.04 | |
| markers: "(daily or not daily) and (remote or not remote) and gpu and (doctest or not doctest)" | |
| pytest_command: "coverage run -m pytest" | |
| composer_package_name: "mosaicml" | |
| - name: "gpu-3.10-2.1" | |
| container: mosaicml/pytorch:2.1.2_cu121-python3.10-ubuntu20.04 | |
| markers: "(daily or not daily) and (remote or not remote) and gpu and (doctest or not doctest)" | |
| pytest_command: "coverage run -m pytest" | |
| composer_package_name: "mosaicml" | |
| - name: "gpu-3.10-2.2" | |
| container: mosaicml/pytorch:2.2.0_cu121-python3.11-ubuntu20.04 | |
| markers: "(daily or not daily) and (remote or not remote) and gpu and (doctest or not doctest)" | |
| pytest_command: "coverage run -m pytest" | |
| composer_package_name: "mosaicml" | |
| name: ${{ matrix.name }} | |
| if: github.repository_owner == 'mosaicml' | |
| with: | |
| composer_package_name: ${{ matrix.composer_package_name }} | |
| container: ${{ matrix.container }} | |
| git_repo: mosaicml/composer | |
| mcloud-timeout: 3600 | |
| name: ${{ matrix.name }} | |
| pip_deps: "[all]" | |
| pytest-command: ${{ matrix.pytest_command }} | |
| pytest-markers: ${{ matrix.markers }} | |
| python-version: 3.9 | |
| secrets: | |
| mcloud-api-key: ${{ secrets.MCLOUD_DAILY_API_KEY }} |