Testing¶

Run the test suite from the repository root:

python -m pytest -q

The default CI path runs CPU-safe tests and skips CUDA-only tests:

python -m pytest -q -m "not cuda"

For a quick environment check, collect tests first:

python -m pytest --collect-only -q

Coverage¶

To run tests with coverage:

python -m pytest -q --cov=torchkm --cov-report=term-missing:skip-covered --cov-report=xml --cov-report=html

Current coverage: 98.00% on commit 29fbdf4, using an NVIDIA L40S.

Test categories¶

Useful test categories include:

estimator interface tests;
low-level solver tests;
CPU fallback tests;
GPU tests when CUDA is available;
input validation tests;
label convention tests;
probability calibration tests;
Nyström approximation tests;
reproducibility tests with fixed random seeds.

CUDA tests¶

CUDA smoke tests are marked with pytest.mark.cuda and skip automatically when CUDA is unavailable. GitHub-hosted runners usually may not have CUDA GPUs, so run the CUDA smoke tests on a CUDA machine with:

python -m pytest -q -m cuda

Small tests are better¶

Keep unit tests small and fast. Large benchmark tests should live in benchmarks/, not in the normal test suite.

Documentation build¶

The documentation build should be lightweight and should not run examples or benchmarks:

mkdocs build --strict