Welcome to ramp's documentation!
Ramp is a python package for rapid machine learning prototyping. It provides a simple, declarative syntax for exploring features, algorithms and transformations quickly and efficiently. At its core it's a unified pandas-based framework for working with existing python machine learning and statistics libraries (scikit-learn, rpy2, etc.).
- Easy and methodical model comparison and evaluation
- Clean components for features, models, and results
- Comprehensive suite of tools for measuring, validating, and building, blending, and composing models.
- Easily extensible API