Using whylogs with MLflow

Describes how to use whylogs with MLflow.

Prerequisites

Sign in to HPE Ezmeral Unified Analytics Software as a member.

About this task

In HPE Ezmeral Unified Analytics Software, whylogs is integrated with MLflow to log and analyze the data quality. You can use whylogs to analyze the data quality throughout the machine learning lifecycle.

To use whylogs with MLflow, refer to the MLflow logging example in the GitHub. The basic steps are outlined as follows:
  1. Create a notebook or import the notebook into HPE Ezmeral Unified Analytics Software. See Creating and Managing Notebook Servers.
  2. Import the required libraries and modules from whylogs.
  3. Train a model and create data frames to profile the data, and then run the notebook.
  4. Once you finish running your notebook, navigate back to the HPE Ezmeral Unified Analytics Software home screen.
  5. Click the Tools & Frameworks icon on the left navigation bar. Navigate to the MLflow tile under the Data Science tab and click Open.
  6. View the whylogs output in the whylogs directory within that run’s artifacts in the MLflow UI.

Results

You can analyze the data quality metrics and ensure the data quality by using whylogs ouput.