Closed Dai-Wenxun closed 3 weeks ago
This issue can be fixed as follows:
self.lengths.append(motion.shape[0] - opt.window_size)
self.lengths.append(motion.shape[0] - opt.window_size + 1) # Fixed
Cool! Thanks for your careful fix. To keep the code consistent, I might not be able to update this.
In the getitem method of the MotionDataset, I found that the LAST frame of each motion segment cannot be retrieved.
For example, in the first motion,
self.cumsum[1] = 129
. Whenitem=129
, there isidx = 128
sinceidx=item-self.cumsum[motion_id=0]-1
, resulting in the motion segment being[128:128+64] = [128:192]
, while in fact, the length of this motion should be129+64=193
according to this line. This issue occurs with every subsequent motion segment. Should I submit a PR to fix this problem?