DOCSEARCHbyAlgolia
Submit Docs

MlFinLab Documentation

repository·master·Indexed 26 days ago

https://github.com/hudson-and-thames/mlfinlab

A professional-grade Python library for financial machine learning research. MlFinLab provides a comprehensive toolbox for the strategy development lifecycle, including data labeling, feature engineering, modeling, backtesting, and overfitting prevention. It includes specialized modules for clustering, cross-validation, hyper-parameter tuning, bet sizing, and synthetic data generation.

Tokens
1.2K
Snippets
2
Records
9
Agent score
89%

What's inside MlFinLab

  1. Overview of MlFinLab

    master
    MlFinLab is a Python library designed for financial machine learning research. It provides a comprehensive toolbox covering the entire ML strategy creation pipeline, including data structure generation, feature engineering, modeling, and backtest statistics. The library is designed for robustness with extensively tested and documented modules.

  3. Review the Non-Commercial License terms and limitations

    master

    The codebase is provided "AS IS" under a Non-Commercial license. Key legal terms include:

    • No Warranties: The Licensor makes no warranties regarding the codebase, its suitability for a particular purpose, or its compatibility with hardware/software. Users assume all risk for results obtained.
    • Limited Liability: The Licensor is not liable for any damages (including loss of profits, business interruption, or data loss) arising from the use or inability to use the software.
    • Audit Rights: The Licensor reserves the right to examine or audit the codebase to check for unauthorized use or modifications.
    • Indemnification: Users must indemnify the Licensor against claims arising from the use of the codebase.
    • Governing Law: This agreement is governed by English law.

  4. Licensing and Support for MlFinLab

    master

    MlFinLab is available under Business and Enterprise licensing options.

    Support and Community:

    • Slack Community: Clients with a purchased license gain access to the Hudson & Thames Slack community for direct interaction with engineers and quants.
    • Email Support: You can contact the research team at research@hudsonthames.org.

  5. Explore MlFinLab modules and learning resources

    master

    MlFinLab includes specialized modules for various stages of the financial machine learning pipeline. For each technique, the library provides documentation (theoretical and functional), lecture videos, slides, and example notebooks that demonstrate full pipelines from data import to performance metrics.

    Included Modules:

    • Backtest Overfitting Tools
    • Data Structures
    • Labeling
    • Sampling
    • Feature Engineering
    • Models
    • Clustering
    • Cross-Validation
    • Hyper-Parameter Tuning
    • Feature Importance
    • Bet Sizing
    • Synthetic Data Generation
    • Networks
    • Measures of Codependence
    • Useful Financial Features

  6. Understand the Non-Commercial License terms

    master

    The MlFinLab codebase is not open-source and is subject to a proprietary non-commercial license. By installing, accessing, or using the codebase, you agree to these terms.

    Key Usage Terms:

    • Permitted Use: You may use the unmodified codebase for research and training purposes in quantitative finance and machine learning.
    • Prohibited Use: Use for any commercial purposes is strictly prohibited under this license. If you require commercial use, you must purchase a Business or Enterprise license from Hudson and Thames Quantitative Research.
    • No Derivatives: You may not create derivative works based on the codebase, nor may you use the source or binary code to reverse engineer the proprietary algorithms.
    • No Redistribution: You cannot sell, rent, lease, loan, or distribute the codebase or any part of it to third parties.
    • No Reverse Engineering: You may not decompile, disassemble, or otherwise attempt to reduce the codebase to human-readable form.

    Ownership and Modifications:

    • The codebase is the proprietary trade secret of Hudson and Thames.
    • Any improvements, enhancements, or modifications made to the codebase (even by the user) automatically become the proprietary property of the Licensor.
    • Users are required to immediately disclose any improvements or modifications made to the codebase to the Licensor.