Implement Router Z-loss #99
Workflow file for this run
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: PR GPU tests | |
| on: | |
| push: | |
| branches: | |
| - main | |
| - release/* | |
| pull_request_target: | |
| branches: | |
| - main | |
| - release/** | |
| workflow_dispatch: | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }} | |
| cancel-in-progress: ${{ github.ref != 'refs/heads/main'}} | |
| jobs: | |
| pytest-gpu: | |
| name: ${{ matrix.name }} | |
| if: github.repository_owner == 'databricks' | |
| runs-on: ubuntu-latest # TODO: switch to linux-ubuntu-latest later | |
| strategy: | |
| fail-fast: false | |
| matrix: | |
| include: | |
| - name: "python3.11-pytorch2.4.0-gpus1" | |
| gpu_num: 1 | |
| python_version: 3.11 | |
| container: mosaicml/pytorch:2.4.0_cu124-python3.11-ubuntu20.04 | |
| - name: "python3.11-pytorch2.4.0-gpus2" | |
| gpu_num: 2 | |
| python_version: 3.11 | |
| container: mosaicml/pytorch:2.4.0_cu124-python3.11-ubuntu20.04 | |
| steps: | |
| - name: Run PR GPU tests | |
| uses: mosaicml/ci-testing/.github/actions/[email protected] | |
| with: | |
| name: ${{ matrix.name }} | |
| container: ${{ matrix.container }} | |
| python_version: ${{ matrix.python_version }} | |
| gpu_num: ${{ matrix.gpu_num }} | |
| git_repo: databricks/megablocks | |
| pip_deps: "[all,testing]" | |
| pytest_command: "coverage run -m pytest tests" | |
| # TODO: remove tests from pytest tests when we delete all tests in the MegaBlocks dir | |
| pytest_markers: "gpu" | |
| composer_package_name: mosaicml # Required as Composer is built from source | |
| mcloud_timeout: 3600 | |
| mcloud_api_key: ${{ secrets.MCLOUD_API_KEY }} |