pretrained models

[xiaodongpo]

Hello，could you please tell me how to use the pretrained models(flatnet_separable_pointGrey_randomInit or flatnet_separable_pointGrey_transposeInit) to train FlatNet-sep. Thank you!

[xiaodongpo]

run main.py to train from scratch may not need the pretrain models, but when i training, there is error: ../aten/src/ATen/native/cuda/Loss.cu:95: operator(): block: [0,0,0], thread: [2,0,0] Assertion target_val >= zero && target_val <= one failed.
Traceback (most recent call last):
File "main.py", line 156, in
disc_err = train_discriminator_epoch(gen, dis, optim_dis, dis_criterion, train_loader, opt.disPreEpochs, disc_err, device)
File "/root/autodl-tmp/Flatnet/flatnet-flatnet-sep/fns_all.py", line 36, in train_discriminator_epoch
dis_loss = criterion(dis(high_res_real), target_real) + criterion(dis(Variable(high_res_fake.data)), target_fake)
File "/root/miniconda3/envs/flatnet/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1553, in _wrapped_call_impl
return self._call_impl(*args, **kwargs)
File "/root/miniconda3/envs/flatnet/lib/python3.8/site-packages/torch/nn/modules/module.py", line 1562, in _call_impl
return forward_call(*args, **kwargs)
File "/root/miniconda3/envs/flatnet/lib/python3.8/site-packages/torch/nn/modules/loss.py", line 621, in forward
return F.binary_cross_entropy(input, target, weight=self.weight, reduction=self.reduction)
File "/root/miniconda3/envs/flatnet/lib/python3.8/site-packages/torch/nn/functional.py", line 3172, in binary_cross_entropy
return torch._C._nn.binary_cross_entropy(input, target, weight, reduction_enum)