Questions about the short-term variations and long-term variations representation.

[tiketiskte]

Dear author, in the main text, you explained that for the obtained two-dimensional tensor X_i^{2D} ∈ R^{f_i×p_i} , each row represents short-term changes and each column represents long-term changes. I don't quite understand the description of each row and column. If a time series t is 336 in length and is decomposed into 2 dimensions of 1*336, does that mean that only 1 long-term change is represented each number? 336 numbers indicating short-term changes?