On conversion shows SpatialFullConvolution has no attribute bias

[aszenz]

The same error is thrown by the load_lua method, the model I'm trying to convert is the DGAN CPU pretrained model from here.

The full error message:
Traceback (most recent call last): File "./convert_torch_2_pytorch.py", line 314, in <module> torch_to_pytorch(args.model,args.output) File "./convert_torch_2_pytorch.py", line 303, in torch_to_pytorch lua_recursive_model(model,n) File "./convert_torch_2_pytorch.py", line 114, in lua_recursive_model copy_param(m,n) File "./convert_torch_2_pytorch.py", line 45, in copy_param if m.bias is not None: n.bias.data.copy_(m.bias) AttributeError: 'SpatialFullConvolution' object has no attribute 'bias'