scientific-ml-classification

What problem does it solve?

Standardizes evaluation of multiple classification models using StratifiedKFold cross-validation to ensure fair comparisons and robust metrics.

Core Features & Use Cases

• Supports multiple classifiers (Logistic Regression, Random Forest, Gradient Boosting, SVM, XGBoost) and evaluates them with ROC-AUC, accuracy, precision, recall, F1, and confusion matrices.
• Integrates OpenML data retrieval for benchmark datasets and reproducible experiments.
• Provides a structured workflow from model definition to metrics reporting and result visualization.

Quick Start

Run a cross-validated comparison of multiple classifiers on a dataset and output ROC-AUC and confusion matrices.