unconditional-time-series-diffusion icon indicating copy to clipboard operation
unconditional-time-series-diffusion copied to clipboard

Adding dynamic real features for TSDiffCond

Open ClaraGrthns opened this issue 1 year ago • 0 comments

Thanks for sharing the code base!

I have a question regarding the class of TSDiffCond. It takes num_feat_dynamic_real as an input but in the _extract_features no dynamic real features are added to the used features. I am working with energy data (wind and solar) and would like to use additional past and future covariates (weather forecasts) to improve my predictions on the respective energy target time series. I was wondering: How you would suggest incorporating those into the code?

ClaraGrthns avatar Jul 12 '24 09:07 ClaraGrthns