How Pilz Works

Pilz is an experimental classification library that takes a different approach to machine learning. Instead of optimizing a loss function through gradient descent, it directly evaluates feature combinations by their ability to separate target from non-target, with all data processing running through SQL.

Core Concepts

• Feature Categorization
• Multi-Dimensional Splits
• SQL-Native Architecture
• Three-Way Splits
• Downsampling
• Imbalanced Data