Skip to content

Fix/interaction plot violin points #904

Fix/interaction plot violin points

Fix/interaction plot violin points #904

Workflow file for this run

name: CI
on:
push:
branches:
- master
- develop
pull_request:
types: [opened, synchronize, reopened]
workflow_dispatch:
concurrency:
# Cancel older runs for the same PR/branch when new commits arrive.
group: ${{ github.workflow }}-${{ github.event.pull_request.number || github.ref }}
cancel-in-progress: true
permissions:
contents: read
jobs:
code_quality:
name: Code Quality
runs-on: ubuntu-latest
timeout-minutes: 20
steps:
- name: Git checkout
uses: actions/checkout@d23441a48e516b6c34aea4fa41551a30e30af803 # v6.1.0
with:
fetch-depth: 0 # required for setuptools-scm to read git tags
persist-credentials: false
# Oldest supported Python: mypy checks against the installed libraries' stubs, which follow the
# interpreter, so code that only works on newer Pythons is caught here.
- name: Set up uv and Python 3.11
uses: astral-sh/setup-uv@1c37ad07a6a961277cf70c0d37d6f313000f5884 # v10.3.0
with:
python-version: "3.11"
activate-environment: true
enable-cache: true
cache-dependency-glob: uv.lock
# Locked environment: `dev` + `test` groups (`default-groups`), the latter because mypy needs the
# optional dependencies that ship types. `--locked` also fails when uv.lock is out of date.
- name: Install dev dependencies
run: uv sync --locked
- name: Check linting
run: make lint
- name: Check formatting
run: make format
- name: Check type annotations
run: make typecheck
# Not a dependency of the test jobs: run zizmor online audits on the workflows.
workflow_audit:
name: Workflow Audit
runs-on: ubuntu-latest
timeout-minutes: 10
steps:
- name: Git checkout
uses: actions/checkout@d23441a48e516b6c34aea4fa41551a30e30af803 # v6.1.0
with:
persist-credentials: false
- name: Set up uv and Python 3.11
uses: astral-sh/setup-uv@1c37ad07a6a961277cf70c0d37d6f313000f5884 # v10.3.0
with:
python-version: "3.11"
activate-environment: true
enable-cache: true
cache-dependency-glob: uv.lock
# Only the `dev` group, without the project: zizmor stays pinned by uv.lock.
- name: Install zizmor
run: uv sync --locked --only-group dev
- name: Audit workflows
env:
GH_TOKEN: ${{ github.token }} # enables zizmor's online audits (e.g. known-vulnerable actions)
run: make workflows
test:
name: Test (Python ${{ matrix.python-version }})
needs: code_quality
runs-on: ubuntu-latest
timeout-minutes: 60
strategy:
fail-fast: false
matrix:
python-version: ["3.11", "3.12", "3.13", "3.14"]
include:
# One job measures coverage and uploads it: the others run the same suite.
- python-version: "3.12"
coverage: true
steps:
- uses: actions/checkout@d23441a48e516b6c34aea4fa41551a30e30af803 # v6.1.0
with:
fetch-depth: 0 # required for setuptools-scm to read git tags
persist-credentials: false
- name: Set up uv and Python ${{ matrix.python-version }}
uses: astral-sh/setup-uv@1c37ad07a6a961277cf70c0d37d6f313000f5884 # v10.3.0
with:
python-version: ${{ matrix.python-version }}
activate-environment: true
enable-cache: true
cache-dependency-glob: uv.lock
# Full suite, NLP included, pinned to uv.lock through a constraints file. Not `uv sync`: uv.lock
# holds the PyPI torch wheel, which drags in several GB of CUDA libraries the tests never use.
# `--torch-backend cpu` installs the same torch version from the CPU-only index instead.
- name: Install dependencies
run: |
uv export --locked --no-hashes --no-emit-project --extra nlp --group test -o constraints.txt
uv pip install --torch-backend cpu -c constraints.txt '.[nlp]' --group test
# NLP test modules skip (not fail) when their imports are missing, so fail loudly here instead.
- name: Check NLP stack is importable
run: python -c "import torch, transformers, sentence_transformers, captum; print(torch.__version__)"
# Models the NLP tests download from the Hugging Face Hub: reuse them instead of hitting the Hub
# from every job. The key follows the tests, so a newly referenced model gets saved too.
- name: Cache Hugging Face models
uses: actions/cache@55cc8345863c7cc4c66a329aec7e433d2d1c52a9 # v6.1.0
with:
path: ~/.cache/huggingface
key: hf-models-${{ hashFiles('tests/**/*.py') }}
restore-keys: hf-models-
- name: Test
if: ${{ !matrix.coverage }}
run: make test
- name: Test and coverage
if: ${{ matrix.coverage }}
run: make coverage-report
- name: Upload coverage to Codecov
if: ${{ matrix.coverage }}
uses: codecov/codecov-action@fb8b3582c8e4def4969c97caa2f19720cb33a72f # v6.0.2
with:
token: ${{ secrets.CODECOV_TOKEN }}
files: ./coverage.xml
fail_ci_if_error: false
disable_search: true
# Pure-Python package: building it once is enough.
- name: Build python package
if: ${{ matrix.coverage }}
run: uv build
docs:
name: Docs
runs-on: ubuntu-latest
timeout-minutes: 20
steps:
- uses: actions/checkout@d23441a48e516b6c34aea4fa41551a30e30af803 # v6.1.0
with:
fetch-depth: 0 # required for setuptools-scm to read git tags
persist-credentials: false
# Same Python as Read the Docs (.readthedocs.yml)
- name: Set up uv and Python 3.11
uses: astral-sh/setup-uv@1c37ad07a6a961277cf70c0d37d6f313000f5884 # v10.3.0
with:
python-version: "3.11"
activate-environment: true
enable-cache: true
cache-dependency-glob: uv.lock
# Same install as Read the Docs. Non-editable, so the docs import the built package, not the checkout.
- name: Install doc dependencies
run: uv sync --locked --no-editable --no-default-groups --group doc --extra report
- name: Prepare tutorials and demo report
run: python scripts/prepare_docs.py
# Any warning fails the build (`--keep-going` still reports them all)
- name: Build docs
run: sphinx-build -W --keep-going -b html docs docs/_build/html
lowest_deps:
name: Lowest Dependencies
needs: code_quality
runs-on: ubuntu-latest
timeout-minutes: 60
steps:
- uses: actions/checkout@d23441a48e516b6c34aea4fa41551a30e30af803 # v6.1.0
with:
fetch-depth: 0 # required for setuptools-scm to read git tags
persist-credentials: false
# Oldest supported Python: the oldest dependency releases may ship no wheels for newer ones.
- name: Set up uv and Python 3.11
uses: astral-sh/setup-uv@1c37ad07a6a961277cf70c0d37d6f313000f5884 # v10.3.0
with:
python-version: "3.11"
activate-environment: true
enable-cache: true
cache-dependency-glob: pyproject.toml
# Each direct dependency at the lower bound pyproject.toml declares (transitive ones at their
# newest), so a PR relying on a newer API fails here until the bound is raised. Classical install:
# the extras the `test` group pulls are listed explicitly so they count as direct too.
- name: Install dependencies
run: |
uv pip install --resolution lowest-direct '.[xgboost,catboost,lightgbm,lime,report]' --group test
python -c "import pandas, plotly, sklearn; print(pandas.__version__, plotly.__version__, sklearn.__version__)"
- name: Check NLP stack is absent
run: python -c "import importlib.util as u; assert u.find_spec('torch') is None, 'torch installed'"
- name: Test
run: make test