Using One AI FAQs

Why are the core attributes not generating in the machine learning model recipe?
I want the model to be able to select from a different scope of columns or range of tables. How do I do that?
Can you exclude the unique identifier column from a machine learning model?
Can you change the null drop threshold for a single column?
Why is my machine learning model erroring after conducting overrides and/or performing advanced configuration?
Can I conduct model configuration and overrides (adjust global settings, dimensionality reduction, upsampling, etc.) before creating the model in the One AI Recipe?
What is the default category size threshold in the global settings for machine learning models?
What is the default null drop threshold in the global settings for machine learning models?
What does One AI mean when it marks a column as a “suspicious” variable on the EDA Report? How does this differ from "leakage"? What are the default thresholds?
Can I force One AI to select a specific column or feature in a model? Can I force One AI drop/exclude a specific column from a model?
How do I make a model select more (or less) features / variables?
Which estimator is best for my machine learning model?
What is upsampling? When should it be leveraged in a machine learning model?
Can I make overrides or perform manual configurations in One AI when creating or updating a machine learning model? What overrides and configurations are available?
Can I rename a machine learning model or data augmentation after creating it?
Why don’t all my metrics show up in the metric drop down when I am creating generative attributes in a machine learning model?
What are generative attributes? How are they different from standard features / attributes?
How can I strengthen my machine learning model?
What is considered an acceptable or good F1 score for machine learning models?
What is F1, precision, and recall and what do the scores tell us? Is the F1 score the most important?
In a binary classification machine learning model, what is "the positive label"?
What is the difference between "Feature Importances" and "SHAP Averages" on the Results Summary Report for machine learning models?
What is the SHAP beeswarm chart? How can I best interpret this chart?
What is SHAP? Why do we use it in One AI and on One AI Storyboards?
What is the EDA report for machine learning models?
Why didn't my machine learning model save the edits I made in the One AI Recipes screen?
Why isn’t the new data from my machine learning model showing in the storyboard when I just deployed the model?
What is the difference between deploy and deploy and persist model for machine learning models?
Can I schedule a machine learning model or data augmentation to rerun on a specific cadence?
Can I model One AI data from machine learning models and/or data augmentations onto Storyboard(s)?