Firstly, thanks for the great work and open source code.
At the data/keypoint.py file, you have manually set the dataset length as 4000. It limits training set to 4000 image pairs, and I believe this should be changed to return self.size.
def __len__(self):
if self.opt.phase == 'train':
return 4000
Maybe I'm missing something. Is there a reason for limiting the dataset?
Firstly, thanks for the great work and open source code.
At the
data/keypoint.py
file, you have manually set the dataset length as 4000. It limits training set to 4000 image pairs, and I believe this should be changed toreturn self.size
.Maybe I'm missing something. Is there a reason for limiting the dataset?