Auto-PyTorch vs LightAutoML

Auto-PyTorch is a free, self-hosted machine-learning toolkit based on PyTorch. It automates algorithm selection and hyperparameter tuning, including neural-network architecture and training settings, within a user-defined search budget. Documented search methods include Bayesian optimisation, meta-learning, SMAC and Hyperband. Supported work includes tabular classification, tabular regression and time-series forecasting. Tabular preprocessing can perform imputation, categorical encoding, scaling and feature processing, while related hyperparameters are tuned. The toolkit can build ensembles from validation-set predictions, with controls for ensemble size and candidate limits. Users can set estimator memory limits and total wall time. Parallel Bayesian optimisation uses Dask.distributed; workers need shared filesystem access to training data and models. Installation documentation specifies Linux, Python 3.7 or later, a C++11-capable compiler and SWIG 3.0. SWIG 4.0 or later is unsupported. Forecasting needs extra dependencies. A Docker image and PyPI installation are documented, under the 3-clause BSD license.

LightAutoML is an open-source Python library for automated machine learning on tabular and text data. It supports binary and multiclass classification and regression, using ready-made tabular, text and WhiteBox presets or modular custom pipelines. Pipeline creation can automate data processing and typing, feature selection, hyperparameter tuning, time utilization and report creation. Documented model classes include linear models, LightGBM and CatBoost boosted trees, neural networks and a WhiteBox scorecard. Its basic tutorial says the library can handle missing values and outliers automatically. Optional installation extras cover NLP, computer vision and reports, and a tutorial demonstrates a SQL data source. Installation is through PyPI with `pip install lightautoml`; the project is licensed under Apache License, Version 2.0. The maker describes it as a way for data scientists and analysts to reduce routine data preparation and model selection work. The current package handles datasets with independent samples in each row; multitable datasets and sequences remain in progress.