MS (multivariate -> univariate)?
Closed this issue · 2 comments
kailu22 commented
Thanks for the amazing work. I wonder if BasicTS provides the option of MS (multivariate -> univariate). I appreciate your time and help.
zezhishao commented
I guess it depends on your model design.
The input and output of the model in BasicTS are [B, L, N, C], where B is the batch size, L is the length of the history time series, N is the number of variables, and C is the number of features (the first dimension is the target value, and the subsequent time variables (TiD, DiW, etc.)).
As long as N time series are processed in parallel, it could be considered univariate.
kailu22 commented
That's very helpful. Thank you so much!