MLflow is an open-source machine learning lifecycle platform that manages the end-to-end process of testing and deploying models. It’s the first of it kind that is open source and it rivals Facebook FBLearner and Uber Michaelangelo. 

In the early days, Facebook encountered several challenges in the ML process, from experimentation to testing and deployment. It was a very complex process that required highly skilled engineers and data scientists to conduct experiments. And when the same data was fed into the same models, the results would differ. 

FBLearner solved these problems. Today, non-data scientists can conduct thousands of experiments and reproduce the exact same results when the same dataset and tools are used. However, this tool belongs to Facebook. Databricks developed MLflow as an open-source alternative that eases the process of testing, experimentation, training, and deployment. 

Project Background

• Tool: MLflow
• Author: Databricks 
• Initial Release: June 2018
• Type: ML lifecycle platform 
• License: Apache-2.0 
• Supports: Multiple frameworks and tools: TensorFlow, PyTorch, XGBoost, etc. 
• Github: mlflow
• Runs On: Cloud and on-prem
• Hardware: GPUs and CPUs
• Twitter: MLflow

Features

• Track and manage experiments
• Reproducible runs
• Light-weight API
• CI/CD workflow integration
• Artifact store
• Versioning