Change theme
Help
Press space for more information.
Show links for this issue (Shortcut: i, l)
Copy issue ID
Previous Issue (Shortcut: k)
Next Issue (Shortcut: j)
Sign in to use full features.
Vote: I am impacted
Notification menu
Refresh (Shortcut: Shift+r)
Go home (Shortcut: u)
Pending code changes (auto-populated)
View issue level access limits(Press Alt + Right arrow for more information)
Request for new functionality
View staffing
Description
-Test dataset features --> In AutoML’s “Evaluate” tab, you can see how many rows were allocated to the test dataset, and you can export predictions from the test dataset to BigQuery. This does not apply to AI Platform Unified.
-Granularity of display --> When looking at results in AutoML, you can see Accuracy/Precision/Recall/FP as percentage and as a fraction (e.g. Recall=36/507). In contrast, in AIPU, fractions are missing, and Accuracy is not shown at all.
-Confusion Matrix --> The confusion matrix in AutoML gives more detail when hovering the mouse , compared to confusion matrix in AIPU.
--Confusion Matrix --> It would be nice if the confusion matrix was updated interactively, when changing the confidence threshold in the "Evaluate" tab.
Customers can reproduce these features by scripting, but it would be desirable to have it automatically implemented as they already exist in AutoML.