Machine Learning Lifecycle and MLflow basics

Tracking and logging you ML models, metrics and parameters.
Explore the docs »

View Demo · Report Bug ·

Table of Contents

• About the Project
• Getting Started
  • Prerequisites
  • Installation
• Usage
• License
• Contact
• Acknowledgements

About The Project

Using the open source software MLflow to track and log machine learning models, their paramters, metrics and more!

Getting Started

To get a local copy up and running follow these simple steps.

Prerequisites

Best would be to make a conda envrionment then install, sklearn, mlfow and jupyter notebook

conda install -c conda-forge mlflow 

Installation

1. Clone the repo

git clone https://github.com/SamuelAdamsMcGuire/mlflow_demo.git

2. Install mentioned packages

conda install

Usage

Open the wine_quality.ipynb notebook in jupyter notebooks. Follow the instructions. The notebook takes you through the process of reading data for the chemical feaures of wine and thier respective quality. The quality is what we are trying to predict. You will train a LR model while you log and track the model with MLflow.

Then you will use the MLflow user interface to compare models, filter models, register models and then also call on models from storage to make predictions.

License

Distributed under the MIT License. See LICENSE for more information.

Contact

Samuel Adams McGuire - email - samuelmcguire@engineer.com

Project Link: https://github.com/SamuelAdamsMcGuire/mlflow_demo

Acknowledgements

• Zoltal C. Toth CTO at Datapao
• Szilard Pafka Chied Data Scientist at EP0CH
• MLflow