Hi, I am currently exploring options to automate ML model experimentation. Can Prefect be consider as an alternative to MLFlow for the same. Does its capabilities covers that of an MLFlow?. Thanks!!!

Hey @Amar Verma, I’ve been thinking about this a lot myself. My current opinion is that it depends what you use MLFlow for. If it is for experiment tracking, artifact registry or model deployment (after testing metrics), you can also do these completely in Prefect. But also note Prefect doesn’t have the optimized views for comparing metrics across experiments. MLFlow has UI views that let you see model performance versus hyperparameter value. You kinda have to create these yourself in Prefect. But nothing is stopping you from combining the two. Prefect can hit the MLFlow API during your Flow runs.

Thanks a lot @Kevin Kho. This helps me to enough to get started with it.