EnsembleSet Documentation

EnsembleSet generates dataset ensembles by applying a randomized sequence of feature engineering methods to a randomized subset of input features.

Version: 1.0-alpha.23

Overview

EnsembleSet is a Python package designed for generating ensemble datasets through randomized feature engineering. It’s particularly useful for training ensemble machine learning models on tabular data prediction and modeling projects.

Key features:

• Generates multiple dataset variations from a single input dataset

• Applies 11 different feature engineering techniques in random sequences

• Supports both training and testing datasets with minimal data leakage

• Outputs ensembles to HDF5 format for efficient storage

• Uses multiprocessing for parallel dataset generation

User Guide

• Installation
  • Requirements
  • Install from PyPI
  • Install from Source
  • Development Installation
  • Documentation Dependencies
  • Verifying Installation
• Quick Start Guide
  • Basic Usage
  • With Test Data
  • With String Features
  • Understanding Parameters
  • Output Format
  • Reading Generated Datasets
  • Next Steps

Reference

• API Reference
  • DataSet Class
  • Feature Engineering Methods
  • Preprocessing Methods
• Configuration
  • String Encoding Methods
  • Numerical Feature Engineering Methods
  • Parameter Selection
  • Customization
  • Example
• Feature Engineering Catalog
  • String Feature Encoding
  • Numerical Feature Engineering
  • Feature Engineering Pipeline

Indices and tables

• Index

• Module Index

• Search Page