ts-kim / RevIN

RevIN: Reversible Instance Normalization For Accurate Time-series Forecasting Against Distribution Shift
MIT License
272 stars 23 forks source link

Question about Dataset Standardization #13

Open cuongth95 opened 1 year ago

cuongth95 commented 1 year ago

Hi,

I appreciate your paper and code.

Could you confirm if you standardized the datasets (applied z-scores) before using RevIN?

Thank you, Huy

Arnold815 commented 9 months ago

Hi,

I appreciate your paper and code.

Could you confirm if you standardized the datasets (applied z-scores) before using RevIN?

Thank you, Huy

Hi there,

I've got the same question with yours, I read some source code using RevIN and found out that they used the StandardScaler to standardize the data before sending them into the model.

I'm still wondering if there is conflict between these two standardization methods, but doing this can improve the model's performance on my private dataset.