Principle:Interpretml Interpret Synthetic Data
Metadata
| Field | Value |
|---|---|
| Sources | Interpretml_Interpret |
| Domains | Testing, Machine_Learning |
| Updated | 2026-02-07 |
Overview
Synthetic dataset generation creates datasets with known ground-truth feature effects for testing and benchmarking interpretable machine learning models.
Description
The make_synthetic utility function generates synthetic datasets with precisely controlled ground-truth effects including main effects, pairwise interactions, and three-way interactions. The generated data includes feature matrices with specified feature types (continuous, nominal, ordinal) and response vectors computed from known effect functions. This enables rigorous testing of interpretable ML models by comparing learned effects against known ground truth, verifying that models correctly recover the true feature relationships.
Usage
Use synthetic data generation for unit testing, integration testing, and benchmarking of interpretable ML models. The known ground-truth effects enable quantitative validation of model accuracy.