Minotor is an open source software for data and machine learning models monitoring.

Data inconsistency, drift and bias are responsible for most of the model failures in the real world. Minotor helps you solve that problem by tracking inconsistency between your training and production data, detecting bias in your model and performance drops.

Minotor has been built to be easy to handle. Download our Python package, plug your model through a decorator, and you are ready to go !

• I'm new to Minotor 😄 Get started now !
• I want to learn how to use Minotor 🤓 Tutorial
• I'm ready to install Minotor 🚀 Installation
• I would like to contribute 🤗 Thanks ! Check out how to contribute
• I have a question or a suggestion ❓ Send us an email at [email protected]

To monitor your first project, copy the three commands in the cell below and run them in your terminal.

$ pip install minotor
$ minotor iris-example
$ minotor

pip install minotor installs Minotor onto your system.

minotor iris-example creates a sample project using the famous iris dataset.

minotor launches the Minotor dashboard, where you can monitor your data and inspect differences between train and prod.

You can now go to http://localhost:8888 in your favorite browser and play with Minotor !

Head to the Minotor tutorial ➡️

Step-by-step installation guide ➡️

Licensed under the Apache License, Version 2.0. Copy of the license

Complete list here