AutoML UI - Validation method mislabelled or wrong
CB-LTD opened this issue · 3 comments
When creating an AutoML job using the UI and using a user supplied dataset for validation, the completed job run states that the validation method used is 'Monte-Carlo cross validation' under 'Run Summary -> View configuration settings', despite the selected setting.
The validation dataset is stated in the 'Inputs' section.
This leaves some ambiguity as to whether the job run has used the selected validation data, as selected, or used Monte-Carlo validation as it states it has.
It'd be great to get this cleared up, and gain confidence in which validation method is being used.
@CB-LTD
It would be great if you could add a link to the documentation you are following for these steps? This would help us redirect the issue to the appropriate team. Thanks!
This isn't in documentation, it is in the GUI of AzureML in the job run within an experiment (see image below).
Apologies that this is the wrong area to raise this! Please feel free to move this issue or direct me to the correct place to raise this.
@CB-LTD
Thanks for your feedback and your contribution to Azure docs.
Feedback for this repository is moving away from GitHub to a system specific to the Microsoft Learn platform. Issues for this repository will soon be disabled, and additional comments from GitHub will no longer be possible. However, we are now tracking and triaging this issue in the new feedback system.
To learn more about our feedback systems, please see Provide feedback for Microsoft Learn content.