Scope

In the scope of the tutorial

This tutorial adresses the following items:

1. How to include kedro-mlflow capabilities in a Kedro project:

   1. create a new kedro project with updated template

   2. update an existing kedro project

2. Configure mlflow inside a Kedro project

3. Version and track objects during execution with mlflow:

   1. Version parameters inside a Kedro project

   2. Version data inside a Kedro project

   3. (COMING in 0.3.0) Version machine learning models inside a Kedro project

   4. (COMING in 0.3.0) Version metrics inside a Kedro project

   5. Open mlflow ui with project configuratio

This is a step by step tutorial and it is recommended to read the different chapters above order, but not mandatory.

Out of scope of the tutorial

Some advanced capabilities are adressed in the advanced use section:

• saving a kedro pipeline as a mlflow model and serve it

• (COMING in 0.3.0) launching a Kedro project directly with mlflow trhough the MLProject file.