Open AstrayChao opened 4 years ago
file: train.py line: 126
classification_loss, regression_loss = retinanet([data['img'].cuda().float(), data['annot']])
I think data['annot'] should be data['annot'].cuda()
data['annot']
data['annot'].cuda()
What is the result if we do not use .cuda() ?
missing something
file: train.py line: 126
I think
data['annot']
should bedata['annot'].cuda()