PySpect

Home

lists

Frequently asked questions

© 2025 PySpect

Package profile

shap

Summary: A unified approach to explain the output of any machine learning model.
Author: Scott Lundberg <slund1@cs.washington.edu>
Homepage:
Number of releases: 107
First release: 0.1 on 2016-12-01
Latest release: 0.48.0 on 2025-06-12

Releases

PyPI Downloads

Dependencies

Shap has 48 dependencies, 38 of which optional.

Dependencies of shap (48).
Dependency	Optional
cloudpickle	false
numba	false
numpy	false
packaging	false
pandas	false
scikit-learn	false
scipy	false
slicer	false
tqdm	false
typing-extensions	false
catboost	true
causalml	true
datasets	true
gpboost	true
ipython	true
ipywidgets	true
jupyter	true
keras	true
lightgbm	true
lime	true
matplotlib	true
mypy	true
myst-parser	true
nbconvert	true
nbformat	true
nbsphinx	true
ngboost	true
nlp	true
numpydoc	true
opencv-python	true
protobuf	true
pyod	true
pyspark	true
pytest	true
pytest-cov	true
pytest-mpl	true
requests	true
selenium	true
sentencepiece	true
sphinx	true
sphinx_github_changelog	true
sphinx_rtd_theme	true
tensorflow	true
tf-keras	true
torch	true
torchvision	true
transformers	true
xgboost	true

Details