awslabs/autogluon
Autogluon
AutoGluon is an automated machine learning framework designed to optimize model selection and hyperparameter tuning across tabular, text, image, and time series data. It functions as an ensemble learning library and a tabular data prediction engine, aiming to build high-accuracy predictive models without manual algorithm selection.
The framework integrates multimodal machine learning pipelines that combine disparate data types into a single representation using specialized encoders. It also includes a probabilistic time series forecaster that fits multiple statistical and deep learning models to temporal sequences to generate future value ranges.
The system covers broad capability areas including automated hyperparameter optimization and pipeline orchestration. It utilizes multi-layer model stacking and weighted averaging to refine accuracy and reduce variance in predictions.
Features
- Automated Machine Learning Tools - Provides an automated framework for building high-accuracy predictive models without manual hyperparameter tuning or algorithm selection.
- Tabular Predictive Models - Optimizes predictive performance on structured datasets through automated feature engineering and multi-layer model stacking.
- Ensemble Learning - Implements ensemble learning techniques by combining multiple diverse models to reduce variance and improve generalisation.
- Ensemble Learning Libraries - Combines diverse machine learning models through weighting and stacking to reduce variance and improve generalisation.
- Automated ML (AutoML) - An automated machine learning library that optimizes model selection and hyperparameter tuning across diverse data types.
- Multimodal Model Pipelines - Creates machine learning pipelines that combine text, images, and tabular data into a single predictive model.
- Hyperparameter Optimization - Provides automated methods for searching and selecting the best configuration parameters to optimize model performance.
- Multimodal Analytical Pipelines - Provides automated pipelines that integrate text, images, and tabular data into a single representation for predictive modeling.
- Multimodal Machine Learning - Combines text, images, and tables into a single automated machine learning pipeline to process diverse data types.
- Multimodal Representation Learning - Combines disparate data types into a single representation by training specialized encoders for text, images, and tables.
- Time Series Forecasting - Fits multiple statistical and deep learning models to temporal sequences for probabilistic time series forecasting.
- Probabilistic Forecasting - Generates future value ranges by fitting multiple statistical and deep learning models to time series data.
- Model Pipeline Orchestration - Orchestrates model pipelines that chain preprocessing and model selection into a consistent workflow.
- Machine Learning Pipelines - Sequences data preprocessing, model selection, and evaluation steps into a repeatable, automated machine learning workflow.
- Encoder-Combiner Architectures - Employs encoder-combiner architectures to map text, images, and tables into a single fused representation.
- Hierarchical Stacking - Utilizes multi-layer model stacking to train new models on previous predictions for refined accuracy.
- Machine Learning Training - Tunes machine learning and deep learning models for diverse data types to maximize predictive accuracy.
- Automated Machine Learning - Automated machine learning for diverse data types.
- AutoML - Automated machine learning for multi-modal data.
- AutoML and Specialized AI - AutoML for image, text, and tabular data.
- AutoML Frameworks - AutoML framework for text, image, and tabular data.
- General Machine Learning - AutoML for image, text, and tabular data.
- MLOps and Pipelines - AutoML for tabular, image, and text data.
- Data Science Tools - AutoML for multi-modal data predictions.
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.
