rhiever/tpot
Tpot
This is a Python automated machine learning framework designed to automate the design and optimization of machine learning pipelines. It functions as a genetic programming pipeline optimizer and an automated feature selection tool, using evolutionary search to discover the most effective sequences of data processing and model steps.
The project focuses on multi-objective optimization to balance competing performance metrics simultaneously. It employs a genetic selection process to identify impactful variables and remove noise from raw datasets, ensuring the resulting machine learning solutions satisfy multiple requirements.
The framework covers a broad capability surface including automated feature engineering, data preprocessing automation, and hyperparameter pipeline tuning. It handles data cleaning tasks such as filling missing values and standardizing features before fitting them to a model.
Features
- Machine Learning Pipelines - Automates the discovery of the optimal sequence of preprocessing and modeling steps for a given dataset.
- Automated Feature Selection Tools - Provides a genetic-based tool to identify impactful variables and remove noise from datasets.
- Feature Engineering - Automates the process of identifying and selecting impactful variables for model input.
- Genetic Algorithms - Uses genetic algorithms to evolve and optimize the structure of machine learning pipelines.
- Genetic Feature Selection - Uses a genetic selection process to identify impactful variables and remove noise from raw data.
- Multi-Objective Tuning - Optimizes multiple competing performance metrics simultaneously to satisfy complex requirements.
- Python Machine Learning Libraries - Provides a Python-based framework for the automated design and optimization of ML pipelines.
- Pipeline Optimizers - Implements a genetic programming system to discover the most effective sequence of data processing and model steps.
- ML Model Compositions - Assembles modular sequences of transformers and estimators into a single executable pipeline.
- Multi-Objective Optimization - Balances multiple competing performance metrics to identify optimal machine learning solutions.
- Fitness-Based Selection - Implements a ranking system to filter high-performing pipeline candidates based on predictive accuracy.
- Hyperparameter Tuning - Searches for the optimal combination of model operators and hyperparameter constraints.
- Data Preprocessing - Provides automated data cleaning, including missing value imputation and feature standardization.
- Search Algorithms - Employs stochastic search and random sampling to explore high-dimensional operator spaces and avoid local optima.
- Search Space Definition - Restricts evolutionary search to a predefined set of compatible ML operators and hyperparameters.
- Automated Machine Learning - Automated machine learning tool that optimizes machine learning pipelines.
- Machine Learning - Automated machine learning tool for pipeline optimization.
- Machine Learning and AI - Automated machine learning tool for optimizing data science pipelines.
Star history
How this analysis was created: This summary and feature list are AI-generated from collected project material and can contain mistakes. Stars, license and language are imported from GitHub. Inclusion does not mean that we have tested or audited this project. Check the source documentation for any feature you depend on. Learn more on our About page.
