name: "BioNemo Image Build and Unit Tests"
on:
push:
branches:
- main
- "pull-request/[0-9]+"
- "dependabot/**"
merge_group:
types: [checks_requested]
schedule:
- cron: "0 7 * * *" # Runs at 7 AM UTC daily (12 AM MST)
defaults:
run:
shell: bash -x -e -u -o pipefail {0}
concurrency:
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: true
jobs:
pre-commit:
runs-on: ubuntu-latest
steps:
- uses: actions/checkout@v4
with:
fetch-depth: 0
submodules: "recursive"
- uses: actions/setup-python@v5
with:
python-version: "3.12"
cache: "pip"
- run: pip install -r requirements-dev.txt
- run: ./ci/scripts/static_checks.sh
# With copy-pr-bot, we need to get the PR labels from the PR API rather than from the event metadata.
get-pr-labels:
runs-on: ubuntu-latest
outputs:
labels: ${{ steps.get-labels.outputs.labels }}
steps:
- name: Get PR number from branch
if: startsWith(github.ref, 'refs/heads/pull-request/')
id: get-pr-num
run: |
PR_NUM=$(echo ${{ github.ref_name }} | grep -oE '[0-9]+$')
echo "pr_num=$PR_NUM" >> $GITHUB_OUTPUT
- name: Get PR labels
id: get-labels
if: startsWith(github.ref, 'refs/heads/pull-request/')
env:
GH_TOKEN: ${{ github.token }}
run: |
LABELS=$(gh api repos/${{ github.repository }}/pulls/${{ steps.get-pr-num.outputs.pr_num }} --jq '[.labels[].name]' || echo "[]")
echo "labels=$LABELS" >> $GITHUB_OUTPUT
- name: Set empty labels for non-PR branches
if: ${{ !startsWith(github.ref, 'refs/heads/pull-request/') }}
id: get-labels-empty
run: echo "labels=[]" >> $GITHUB_OUTPUT
build-bionemo-image:
needs:
- pre-commit
- get-pr-labels
runs-on: linux-amd64-cpu16
if: ${{ !contains(fromJSON(needs.get-pr-labels.outputs.labels || '[]'), 'SKIP_CI') }}
steps:
- name: Login to Docker Hub
uses: docker/login-action@v3
with:
username: ${{ vars.DOCKER_USERNAME }}
password: ${{ secrets.DOCKER_PASSWORD }}
- name: Checkout repository
uses: actions/checkout@v4
with:
submodules: "recursive"
- name: Set up Docker Buildx
uses: docker/setup-buildx-action@v3
- name: Docker Metadata
id: metadata
uses: docker/metadata-action@v5
with:
images: svcbionemo023/bionemo-framework
tags: |
type=schedule
type=ref,event=branch
type=ref,event=tag
type=ref,event=pr
type=raw,value=${{ github.run_id }}
# This action sets up our cache-from and cache-to flags appropriately; see the README of this action for more
# info. It doesn't seem to cache correctly for merge_group events, so we need to add that as an extra argument in
# the step below. There's probably a slight optimization to be had here by caching from the pr- caches for
# merge_group events. See https://github.com/int128/docker-build-cache-config-action/issues/1005 for more info.
- uses: int128/docker-build-cache-config-action@v1
id: cache
with:
image: svcbionemo023/bionemo-build-cache
- name: Build and push
uses: docker/build-push-action@v5
with:
push: true
tags: ${{ steps.metadata.outputs.tags }}
labels: ${{ steps.metadata.outputs.labels }}
cache-from: |
${{ steps.cache.outputs.cache-from }}
type=registry,ref=svcbionemo023/bionemo-build-cache:main
cache-to: ${{ steps.cache.outputs.cache-to }}
run-tests:
needs:
- build-bionemo-image
- get-pr-labels
runs-on: linux-amd64-gpu-l4-latest-1
container:
image: svcbionemo023/bionemo-framework:${{ github.run_id }}
credentials:
username: ${{ vars.DOCKER_USERNAME }}
password: ${{ secrets.DOCKER_PASSWORD }}
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Run tests
# Tests in this stage generate code coverage metrics for the repository
# Coverage data is uploaded to Codecov in subsequent stages
env:
BIONEMO_DATA_SOURCE: ngc
run: ./ci/scripts/run_pytest.sh --no-nbval --skip-slow
- name: Upload coverage to Codecov
# Don't run coverage on merge queue or nightly CI to avoid duplicating reports
# to codecov. See https://github.com/matplotlib/napari-matplotlib/issues/155
if: github.event_name != 'merge_group' && github.event_name != 'schedule'
uses: codecov/codecov-action@v5
with:
token: ${{ secrets.CODECOV_TOKEN }}
- name: Upload test results to Codecov
# Don't run coverage on merge queue or nightly CI to avoid duplicating reports
# to codecov. See https://github.com/matplotlib/napari-matplotlib/issues/155
if: ${{ !cancelled() && github.event_name != 'merge_group' && github.event_name != 'schedule' }}
uses: codecov/test-results-action@v1
with:
token: ${{ secrets.CODECOV_TOKEN }}
run-slow-tests:
needs:
- build-bionemo-image
- get-pr-labels
runs-on: linux-amd64-gpu-l4-latest-1
container:
image: svcbionemo023/bionemo-framework:${{ github.run_id }}
credentials:
username: ${{ vars.DOCKER_USERNAME }}
password: ${{ secrets.DOCKER_PASSWORD }}
if: |
github.event_name == 'schedule' || github.event_name == 'merge_group' ||
contains(fromJSON(needs.get-pr-labels.outputs.labels || '[]'), 'INCLUDE_SLOW_TESTS')
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Run slow tests
env:
BIONEMO_DATA_SOURCE: ngc
# Not every sub-package has slow tests, and since some sub-packages have tests under the same name we need
# to run package by package like we do with the fast tests.
run: ./ci/scripts/run_pytest.sh --no-nbval --only-slow --allow-no-tests
run-notebooks-docs:
needs:
- build-bionemo-image
- get-pr-labels
runs-on: linux-amd64-gpu-l4-latest-1
if: |
github.event_name == 'schedule' || github.event_name == 'merge_group' ||
contains(fromJSON(needs.get-pr-labels.outputs.labels || '[]'), 'INCLUDE_NOTEBOOKS_TESTS')
container:
image: svcbionemo023/bionemo-framework:${{ github.run_id }}
credentials:
username: ${{ vars.DOCKER_USERNAME }}
password: ${{ secrets.DOCKER_PASSWORD }}
steps:
- name: Checkout repository
uses: actions/checkout@v4
- name: Run notebook tests
env:
BIONEMO_DATA_SOURCE: ngc
# this variable should be used in the notebooks to run a subset of the model layers or a smaller model/dataset
FAST_CI_MODE: true
run: |
pytest -v --nbval-lax -p no:python docs/ sub-packages/
verify-tests-status:
# Base on the status of this job, the unit-tests workflow succeeds or fails
# This steps checks the status of all test jobs and fails if any of them failed or were cancelled.
# It is a workaround for the lack of a built-in feature to finalize a pipeline by checking the status of multiple jobs
needs: # List all your run-*-test jobs
- run-tests
- run-slow-tests
- run-notebooks-docs
# Add all other run-*-test jobs
runs-on: ubuntu-latest
if: always() # This ensures the job runs even if previous jobs fail
steps:
- name: Check test job statuses
run: |
if [[ "${{ contains(needs.*.result, 'failure') || contains(needs.*.result, 'cancelled') }}" == "true" ]]; then
echo "Some test jobs have failed or been cancelled!"
exit 1
else
echo "All test jobs have completed successfully or been skipped!"
exit 0
fi
Datasets
Models
