Tools for machine learning model validation
Project description
Probatus
Overview
Developing a well-performing model is often the easy part. What comes next, validating the model, is often overlooked. Probatus is a Python library that standardizes and collects different validation steps to be performed on the models. It focuses on binary classification models to test and measure model robustness.
Installation
You can install probatus using pip:
pip install probatus
Alternatively you can fork/clone and run:
git clone https://gitlab.com/ing_rpaa/probatus.git
cd probatus
pip install .
Dependencies
Probatus requires:
- scikit-learn (>= 0.20.2)
- pandas (>= 0.25)
- matplotlib (>= 3.1.1)
- seaborn (>= 0.9.0)
- shap (>= 0.32)
- scipy (>= 1.4.0)
- joblib (>= 0.13.2)
Documentation
Latest documentation can be found here.
Contribution
To learn more about making a contribution to probatus, please see CONTRIBUTING.md
.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
probatus-0.1.3.tar.gz
(245.5 kB
view hashes)