About LightAutoML

Fast and customizable framework for automatic ML model creation (AutoML)

s

Published by

sb-ai-lab

Visit View Profile

README.md

View on GitHub

Read the Docs

Documentation | Installation | Examples | Telegram chat | Telegram channel

LightAutoML (LAMA) allows you create machine learning models using just a few lines of code, or build your own custom pipeline using ready blocks. It supports tabular, time series, image and text data.

Authors: Alexander Ryzhkov, Anton Vakhrushev, Dmitry Simakov, Rinchin Damdinov, Vasilii Bunakov, Alexander Kirilin, Pavel Shvets.

Quick tour

There are two ways to solve machine learning problems using LightAutoML:

Ready-to-use preset:

  from lightautoml.automl.presets.tabular_presets import TabularAutoML
  from lightautoml.tasks import Task

  automl = TabularAutoML(task = Task(name = 'binary', metric = 'auc'))
  oof_preds = automl.fit_predict(train_df, roles = {'target': 'my_target', 'drop': ['column_to_drop']}).data
  test_preds = automl.predict(test_df).data

As a framework:
LightAutoML framework has a lot of ready-to-use parts and extensive customization options, to learn more check out the resources section.

Resources

Kaggle kernel examples of LightAutoML usage:

Google Colab tutorials and other examples:

Tutorial_1_basics.ipynb - get started with LightAutoML on tabular data.
Tutorial_2_WhiteBox_AutoWoE.ipynb - creating interpretable models.
Tutorial_3_sql_data_source.ipynb - shows how to use LightAutoML presets (both standalone and time utilized variants) for solving ML tasks on tabular data from SQL data base instead of CSV.
Tutorial_4_NLP_Interpretation.ipynb - example of using TabularNLPAutoML preset, LimeTextExplainer.
Tutorial_5_uplift.ipynb - shows how to use LightAutoML for a uplift-modeling task.
Tutorial_6_custom_pipeline.ipynb - shows how to create your own pipeline from specified blocks: pipelines for feature generation and feature selection, ML algorithms, hyperparameter optimization etc.
Tutorial_7_ICE_and_PDP_interpretation.ipynb - shows how to obtain local and global interpretation of model results using ICE and PDP approaches.
Tutorial_8_CV_preset.ipynb - example of using TabularCVAutoML preset in CV multi-class classification task.
Tutorial_9_neural_networks.ipynb - example of using Tabular preset with neural networks.
Tutorial_10_relational_data_with_star_scheme.ipynb - example of using Tabular preset with neural networks.
Tutorial_11_time_series.ipynb - example of using Tabular preset with timeseries data.

Note 1: for production you have no need to use profiler (which increase work time and memory consomption), so please do not turn it on - it is in off state by default

Note 2: to take a look at this report after the run, please comment last line of demo with report deletion command.

Courses, videos

LightAutoML crash courses:
- (Russian) AutoML course for OpenDataScience community
Video guides:
- (Russian) LightAutoML webinar for Sberloga community (Alexander Ryzhkov, Dmitry Simakov)
- (Russian) LightAutoML hands-on tutorial in Kaggle Kernels (Alexander Ryzhkov)
- (English) Automated Machine Learning with LightAutoML: theory and practice (Alexander Ryzhkov)
- (English) LightAutoML framework general overview, benchmarks and advantages for business (Alexander Ryzhkov)
- (English) LightAutoML practical guide - ML pipeline presets overview (Dmitry Simakov)
Articles about LightAutoML:
- (English) LightAutoML vs Titanic: 80% accuracy in several lines of code (Medium)
- (English) Hands-On Python Guide to LightAutoML – An Automatic ML Model Creation Framework (Analytic Indian Mag)

Installation

To install LAMA framework on your machine from PyPI:

# Base functionality:
pip install -U lightautoml

# For partial installation use corresponding option
# Extra dependencies: [nlp, cv, report] or use 'all' to install all dependencies
pip install -U lightautoml[nlp]
# Or extra dependencies with specific version
pip install 'lightautoml[all]==0.4.0'

Additionally, run following commands to enable pdf report generation:

# MacOS
brew install cairo pango gdk-pixbuf libffi

# Debian / Ubuntu
sudo apt-get install build-essential libcairo2 libpango-1.0-0 libpangocairo-1.0-0 libgdk-pixbuf2.0-0 libffi-dev shared-mime-info

# Fedora
sudo yum install redhat-rpm-config libffi-devel cairo pango gdk-pixbuf2

# Windows
# follow this tutorial https://weasyprint.readthedocs.io/en/stable/install.html#windows

Advanced features

GPU and Spark pipelines

Full GPU and Spark pipelines for LightAutoML currently available for developers testing (still in progress). The code and tutorials for:

GPU pipeline is available here
Spark pipeline is available here

Contributing to LightAutoML

If you are interested in contributing to LightAutoML, please read the Contributing Guide to get started.

Support and feature requests

Seek prompt advice in Telegram group.
Open bug reports and feature requests on GitHub issues.

Citation

If you mention LightAutoML in your publications, please cite our paper: Vakhrushev, et al. "LightAutoML: AutoML Solution for a Large Financial Services Ecosystem" arXiv:2109.01528, 2021.

BibTeX entry:

@article{vakhrushev2021lightautoml,
  title={Lightautoml: Automl solution for a large financial services ecosystem},
  author={Vakhrushev, Anton and Ryzhkov, Alexander and Savchenko, Maxim and Simakov, Dmitry and Damdinov, Rinchin and Tuzhilin, Alexander},
  journal={arXiv preprint arXiv:2109.01528},
  year={2021}
}

License

This project is licensed under the Apache License, Version 2.0. See LICENSE file for more details.

Back to top

LightAutoML