Skip to content

Latest commit

 

History

History
31 lines (19 loc) · 1.03 KB

4-tracking-experiments.md

File metadata and controls

31 lines (19 loc) · 1.03 KB

Exercise 4: Tracking Experiments with MLflow

Goals

  • Learn about MLflow
  • Configure our Machine Learning training Pipeline to track experiments and results

Step by Step instructions

  1. In GoCD, click on the little gear symbol (gear) next to ml-pipeline-X to edit your machine learning pipeline configuration.

  2. Open the "Environment Variables" tab and configure the URL to the MLFlow tracking server and your tenant:

  • MLFLOW_TRACKING_URL = http://mlflow.cd4ml.net
  • TENANT = userX (replace X with your user ID)

Configure ML pipeline to track with MLflow

  1. Save and return to the Dashboard page.

  2. Trigger a new ML training pipeline and wait for it to succeed.

  3. Visit the MLflow URL http://mlflow.cd4ml.net to find your experiment results.

Tracking experiment runs with MLflow

  1. Done! Go to the next exercise