The problem of the number of categories in pre-training？

Open Yuyiba opened this issue 5 years ago • 1 comments

Hi,This is perhaps a very simple question, but I don’t understand Using my data set, there are 5 categories,I wonder if we need change 19 to 5? thanks! new_params = model.state_dict().copy() for i in saved_state_dict: # Scale.layer5.conv2d_list.3.weight i_parts = i.split('.') # print i_parts if not args.num_classes == 19 or not i_parts[1] == 'layer5': new_params['.'.join(i_parts[1:])] = saved_state_dict[i] # print i_parts model.load_state_dict(new_params)

Oct 19 '20 00:10 Yuyiba

which file is that in?

Dec 30 '20 07:12 lychrel