generated from AzulGarza/python-project-template
-
-
Notifications
You must be signed in to change notification settings - Fork 51
Open
Labels
questionFurther information is requestedFurther information is requested
Description
Description
Hi guys,
Great job with Timecopilot. You are really pushing very interesting products and libraries in the last years. :)
I have one question/possible suggestion about Timecopilot: from what I have seen, it is not possible to understand how exactly the Agent performs the cross-validation.
Indeed, the cross_validation method of Forecaster has a default value for the windows parameter of 1. This would mean that the cross-validation is performed just on the last h points of the training-set.
However, seeing the logs, it seems clear that indeed the Agent is performing a more thourugh cross-validation.
TLDR:
- can we tune the cross-validation with
analyze? - can we access the "full" results of the cross-validation (i.e., the actual return value of
cross_validationof the Forecaster)
Thanks,
AlexMV12
Context
timecopilot==0.0.20
Metadata
Metadata
Assignees
Labels
questionFurther information is requestedFurther information is requested