Giters

facebookresearch / Kats

Kats, a kit to analyze time series data, a lightweight, easy-to-use, generalizable, and extendable framework to perform time series analysis, from understanding the key statistics and characteristics, detecting change points and anomalies, to forecasting future trends.

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Hyper Parameter Tuning Explanation

F1nalFortune opened this issue · comments

commented

Kats

Describe issue

I just wanted to take a moment to thank the creators of this package for their hard work and dedication. It has been incredibly helpful in my work and I appreciate all the effort that went into making it.

However, I have theory question that is not found within the documentation. When running a hyper-parameter optimization as seen in the picture below, the output returns a column called "mean" - does this mean that the parameter tuning is conducting some sort of k-fold validation? Or is the optimization done only on 1 fold?

Thank you again for all your hard work and dedication, and I look forward to continuing to use this package in my projects.

Provide reproducible example

Example can be seen within the 201 Forecasting Kats Tutorial Notebook.

validation_image