RuntimeError while running CenterNet on visdrone training set on ctdet task CUDA9.0 pytorch 1.1.0

[cnr0724]

When I try to train CenterNet, I get the error like this.

File "main.py" line 102, in
main(opt)
File "main.py", line 70, in main
log_dict_train, _ = trainer.train(epoch, train_loader)
File "/root/CenterNet/src/lib/trains/base_trainer.py", line 119, in train
return self.run_epoch('train', epoch, data_loader)
File "/root/CenterNet/src/lib/trains/base_trainer.py", line 69, in run_epoch
output,loss,loss_stats = model_with_loss(batch)
File "/root/anaconda3/envs/CenterNet/lib/python3.6/site-packages/torch/nn/modules/module.py" , line 493, in call
result = self.forward(*input, **kwargs)
File "/root/CenterNet/src/lib/trains/base_trainer.py", line 19, in forward
outputs= self.model(batch['input'])
File "/root/anaconda3/envs/CenterNet/lib/python3.6/site-packages/torch/nn/modules/module.py" , line 493, in call
result = self.forward(*input, **kwargs)
File "/root/CenterNet/src/lib/models/networks/pose_dla_dcn.py", line 474, in forward
x = self.base(x)
File "/root/anaconda3/envs/CenterNet/lib/python3.6/site-packages/torch/nn/modules/module.py" , line 493, in call
result = self.forward(*input, **kwargs)
File "/root/CenterNet/src/lib/models/networks/pose_dla_dcn.py", line 293, in forward
x = getattr(self, 'level{}'.format(i))(x)
File "/root/anaconda3/envs/CenterNet/lib/python3.6/site-packages/torch/nn/modules/module.py" , line 493, in call
result = self.forward(*input, **kwargs)
File "/root/CenterNet/src/lib/models/networks/pose_dla_dcn.py",line 217, in forward
x1 = self.tree1(x, residual)
File "/root/anaconda3/envs/CenterNet/lib/python3.6/site-packages/torch/nn/modules/module.py" , line 493, in call
result = self.forward(*input, **kwargs)
File "/root/CenterNet/src/lib/models/networks/pose_dla_dcn.py",line 217, in forward
x1 = self.tree1(x, residual)
File "/root/anaconda3/envs/CenterNet/lib/python3.6/site-packages/torch/nn/modules/module.py" , line 493, in call
result = self.forward(*input, **kwargs)
File "/root/CenterNet/src/lib/models/networks/pose_dla_dcn.py", line 59, in forward
out += residual
RuntimeError: The size of tensor a (68) must match the size of tensor b (67) at non-singleton dimension 3

I tried to reshape the tensor to solve the runtimeError but it didn't work. (Another error telling that the shapes of the out tensor and the residual tensor are different occured.) What should I do in order to make it work?

[idantony]

How do you fix this issue?