What is the EDA report for machine learning models?


EDA stands for exploratory data analysis.  One AI creates an EDA report each time it runs a machine learning model and the model is in a status of pending, deployed, or deployed and persisted. EDA is all about understanding the data used by a model to ensure the model can be trusted. A user can use the EDA report to understand correlations, descriptive statistics, and data cleanliness as it relates to the topic at hand. This report will show you what variables the model selected and which ones it didn't select. For variables that were dropped, it will tell you why (i.e., too null, too correlated to another column, etc.). You can analyze variables in this report as well (distinct counts, % missing data, min/max, etc.), visualize your missing data, and download comprehensive correlation data. Check out this EDA help article for a full walkthrough of your EDA reports. 

Navigation: Select One AI from the top navigation bar and Edit on a model

EDA Report example



Was this article helpful?

0 out of 0 found this helpful



Please sign in to leave a comment.