model-explain 0.2.0

A model explainability toolkit with SHAP, LIME, scoring, summaries, and GUI sandbox

Model Explainability Toolkit

This toolkit provides SHAP and LIME-based explanations for scikit-learn models, along with visualization tools.

Features

• SHAP and LIME explainers
• Feature importance plots
• Modular design for easy extension

Installation

Install via pip:

pip install model-explain

API Reference

shap_explainer.shap_explainer(model, X)

Generates SHAP explanations for a fitted scikit-learn model.

• model: Trained scikit-learn estimator
• X: DataFrame of input features

lime_explainer.lime_explainer(model, X)

Generates LIME explanations for a fitted scikit-learn model.

• model: Trained scikit-learn estimator
• X: DataFrame of input features

Example

from model_explain.explainers import shap_explainer
from sklearn.ensemble import RandomForestClassifier
from sklearn.datasets import load_iris
from sklearn.model_selection import train_test_split
import pandas as pd
# Load data
data = load_iris()
X = pd.DataFrame(data.data, columns=data.feature_names)
y = pd.Series(data.target)
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
# Train model
my_model = RandomForestClassifier()
my_model.fit(X_train, y_train)
# Explain model
shap_explainer.shap_explainer(my_model, X_test)

Usage

See examples/demo_notebook.ipynb for a walkthrough.

Support

For questions or issues, open an issue on GitHub or email the maintainer.