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:

• Tabular Playground Series April 2021 competition solution
• Titanic competition solution (80% accuracy)
• Titanic 12-code-lines competition solution (78% accuracy)
• House prices competition solution
• Natural Language Processing with Disaster Tweets solution
• Tabular Playground Series March 2021 competition solution
• Tabular Playground Series February 2021 competition solution
• Interpretable WhiteBox solution
• Custom ML pipeline elements inside existing ones
• Custom ML pipeline elements inside existing ones
• Tabular Playground Series November 2022 competition solution with Neural Networks

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