Getting Started

Installation and Dependencies

You can install PySliceKit using pip:

pip install pyslicekit

PySliceKit relies on standard data science libraries:

• pandas (>= 1.0.0)

• numpy (>= 1.18.0)

• scikit-learn (>= 0.22.0)

• scipy (>= 1.4.0)

• matplotlib (>= 3.2.0)

Supported Metrics

You must pass a valid string to the metric parameter. PySliceKit automatically understands whether higher or lower is better, and automatically selects the correct statistical test for the task type.

Metric string	Task	Direction	Test used
accuracy	Classification	higher is better	Z-test / Fisher
f1, f1_macro, f1_weighted	Classification	higher is better	Z-test / Fisher
precision, recall	Classification	higher is better	Z-test / Fisher
mae, rmse, mse	Regression	lower is better	Bootstrap CI
r2	Regression	higher is better	Bootstrap CI

What it Returns

The pyslicekit.evaluate() function returns a list of SliceResult objects, sorted by absolute gap (worst performing segments first).

You can loop through them or extract the exact properties you need:

for result in results[:5]:  # top 5 worst
    print(f"Segment: {result.label}")
    print(f"Gap: {result.gap:.3f}")
    print(f"Significant: {result.is_significant}")

Complete Minimal Example

import pandas as pd
from sklearn.datasets import load_breast_cancer
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
import pyslicekit

# 1. Load your data and train a model
cancer = load_breast_cancer(as_frame=True)
df = cancer.frame
X = df.drop(columns=['target'])
y = df['target']

X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.25, random_state=42)
model = LogisticRegression(max_iter=5000)
model.fit(X_train, y_train)
y_pred = model.predict(X_test)

# 2. Evaluate!
results = pyslicekit.evaluate(
    model=model,
    df=X_test,
    y_true=y_test,
    y_pred=y_pred,
    slice_cols=["mean radius", "mean texture"],
    metric="f1",
    render_visuals=True,
    top_n=15
)