Skip to content

CONTRIBUTING.md

Latest commit

 

History

History
34 lines (25 loc) · 1.57 KB

CONTRIBUTING.md

File metadata and controls

34 lines (25 loc) · 1.57 KB

Contributing

Thanks for improving scMultiBench. The repository is organized as a benchmark script collection, so contributions should preserve script reproducibility and make assumptions explicit.

Before Opening a Pull Request

  1. Add or update documentation for new inputs, outputs, and dependencies.
  2. Keep generated files out of Git, including __pycache__, .pyc, build/, dist/, and *.egg-info/.
  3. Add or update metadata/methods.yaml when changing method scripts.
  4. Prefer small demo data for examples. Large benchmark data should be linked externally with a checksum or version note.
  5. Run the lightweight checks:
python -m unittest discover -s tests
python scripts/validate_inputs.py classification \
  --reference data/classification/demo_data/data1.h5 \
  --query data/classification/demo_data/data2.h5 \
  --reference-labels data/classification/demo_data/cty1.csv \
  --query-labels data/classification/demo_data/cty2.csv

Adding a Method

For a new method, include:

  • The method script under tools_scripts/<MethodName>/.
  • The exact method version or commit used.
  • The original installation instructions or a method-specific environment file.
  • The expected input format and output format.
  • A registry entry in metadata/methods.yaml.

Dependency Guidance

Avoid forcing all methods into one environment. Many methods depend on incompatible versions of Python, R, PyTorch, TensorFlow, Seurat, scanpy, or scvi-tools. Use the common evaluation environments in envs/ for validation and metrics, and use method-specific environments for full benchmark runs.