pip install shap==0.39.0

A unified approach to explain the output of any machine learning model.

Source
Among top 1000 packages on PyPI.
Over 9.2M downloads in the last 90 days.

Commonly used with shap

Based on how often these packages appear together in public requirements.txt files on GitHub.

lime

Local Interpretable Model-Agnostic Explanations for machine learning classifiers

aix360

IBM AI Explainability 360

osqp

OSQP: The Operator Splitting QP Solver

interpret-core

Fit interpretable machine learning models. Explain blackbox machine learning.

interpret

Fit interpretable machine learning models. Explain blackbox machine learning.

xtoy

get xtoyed predictions from raw data

kaggle

Kaggle API

xport

SAS XPORT file reader

datashader

Data visualization toolchain based on aggregating into a grid

dash-cytoscape

A Component Library for Dash aimed at facilitating network visualization in Python, wrapped around Cytoscape.js

comet-ml

Supercharging Machine Learning

tensorboardX

TensorBoardX lets you watch Tensors Flow without Tensorflow

dash-table

Dash table

skrebate

Relief-based feature selection algorithms

wurlitzer

Capture C-level output in context managers

panel

A high level app and dashboarding solution for Python.

yellowbrick

A suite of visual analysis and diagnostic tools for machine learning.

menpo3d

Menpo library providing tools for 3D Computer Vision research

joblib

Lightweight pipelining: using Python functions as pipeline jobs.

Version usage of shap

Proportion of downloaded versions in the last 3 months (only versions over 1%).

0.39.0

47.05%

0.35.0

30.33%

0.36.0

9.88%

0.37.0

2.90%

0.38.1

2.40%

0.25.2

2.20%

0.34.0

2.06%