Welcome to julearn’s documentation!

… a user-oriented machine-learning library.

What is julearn?

At the Applied Machine Learning (AML) group, as part of the Institute of Neuroscience and Medicine - Brain and Behaviour (INM-7), we thought that using ML in research could be simpler.

In the same way as seaborn provides an abstraction of matplotlib’s functionality aiming for powerful data visualization with minor coding, we built julearn on top of scikit-learn.

julearn is a library that provides users with the possibility of easy testing ML models directly from pandas DataFrames, while keeping the flexibility of using scikit-learn’s models.

To get started with julearn just keep reading here. Additionally you can check out our video tutorial.

Why julearn?

Why not just use scikit-learn? julearn offers three essential benefits:

1. You can do machine learning with less amount of code than in scikit-learn.

2. julearn helps you build and evaluate pipelines in an easy way and thereby helps you avoid data leakage!

3. It offers you nice additional functionality:

   • Easy to implement confound removal: julearn offers you a simple way to remove confounds from your data in a cross-validated way.

   • Data typing: julearn provides a system to specify data types for your features, and then provides you with the possibility to filter and transform your data according to these types.

   • Model inspection: julearn provides you with a simple way to inspect your models and pipelines, and thereby helps you to understand what is going on in your pipeline.

   • Model comparison: julearn provides out-of-the-box interactive visualizations and statistics to compare your models.

Table of Contents

• 1. Getting started
  • 1.1. Requirements
• 2. Setup suggestion
• 3. Installing
• 4. Optional Dependencies
• 5. What you really need to know
  • 5.1. Why cross validation?
  • 5.2. Data
  • 5.3. Model Building
  • 5.4. Model Evaluation
  • 5.5. Model Comparison
• 6. Selected deeper topics
  • 6.1. Applying preprocessing to the target
  • 6.2. Cross-validation consistent Confound Removal
  • 6.3. Hyperparameter Tuning
  • 6.4. Inspecting Models
  • 6.5. Cross-validation splitters
  • 6.6. Stacking Models
  • 6.7. Connectome-based Predictive Modeling (CBPM)
• 7. Overview of available Pipeline Steps
  • 7.1. Transformers
  • 7.2. Models (Estimators)
• 8. Examples
  • 8.1. Starting with julearn
  • 8.2. Model Comparison
  • 8.3. Inspection
  • 8.4. Complex Models
  • 8.5. Confounds
  • 8.6. Customization
• 9. API Reference
  • 9.1. Main API
  • 9.2. Pipeline
  • 9.3. Model Selection
  • 9.4. Base
  • 9.5. Inspect
  • 9.6. Models
  • 9.7. Dynamic Selection (DESLib)
  • 9.8. Scoring
  • 9.9. Scoring Metrics
  • 9.10. Transformers
  • 9.11. Target Transformers
  • 9.12. Utils
  • 9.13. Typing
  • 9.14. Prepare
  • 9.15. Stats
  • 9.16. Visualization
• 10. Configuring julearn
• 11. Contributing
  • 11.1. Setting up the local development environment
  • 11.2. GitHub Pull Request guidelines
  • 11.3. Running unit tests
  • 11.4. Adding and building documentation
  • 11.5. Writing Examples
• 12. Maintaining
  • 12.1. Versioning
  • 12.2. Releasing a new version
• 13. FAQs
  • 13.1. I have issues with the dependencies for the viz module.
  • 13.2. How do I use the viz interactive plots?
• 14. What’s new
  • 14.1. Julearn 0.3.1.dev105 (2024-01-23)
  • 14.2. Julearn 0.2.5 (2022-07-21)

Indices and tables

• Index

• Module Index

• Search Page

Indices and tables

If you use julearn in a scientific publication, please use the following reference

Hamdan, Sami, Shammi More, Leonard Sasse, Vera Komeyer, Kaustubh R. Patil, and Federico Raimondo. ‘Julearn: An Easy-to-Use Library for Leakage-Free Evaluation and Inspection of ML Models’. arXiv, 19 October 2023. https://doi.org/10.48550/arXiv.2310.12568.

Since julearn is also heavily reliant on scikit-learn, please also cite them: https://scikit-learn.org/stable/about.html#citing-scikit-learn