About

This canvas implements a comprehensive end-to-end machine learning pipeline for predicting credit amounts from user event data, featuring modular preprocessing functions, multiple regression models (Ridge, Random Forest, Gradient Boosting), rigorous model evaluation with cross-validation, and publication-ready visualizations using the Zerve design system. The workflow demonstrates best practices in reproducibility, configuration management, and feature importance analysis, with Gradient Boosting achieving 96.8% R² performance on 160K records across 24 engineered features.