TSAM
TSAM copied to clipboard
Implementation about TSAM Conv
Hi, Xueyan
Here is Figure 2 (2) of your paper.

The inputs of the two conv layers are different: one is the shifted features and the other one is the original feature without shifting.
However, in the mode/tsm/temporal_shift.py, the inputs of the two conv layers are the same in line 76.

Could you please help to explain more about this?
Many thanks!