RuntimeError: CUDA error: out of memory

[pyzhang2008]

I was trying to test mode 1~3, while it met CUDA error.

Traceback (most recent call last):
File "test.py", line 388, in
Testing()
File "test.py", line 42, in Testing
lane_agent.load_weights(640, "tensor(0.2298)")
File "/home//PINet/agent.py", line 292, in load_weights
torch.load(self.p.model_path+str(epoch)+''+str(loss)+''+'lane_detection_network.pkl'),False
File "/home//anaconda3/lib/python3.7/site-packages/torch/serialization.py", line 593, in load
return _legacy_load(opened_file, map_location, pickle_module, pickle_load_args)
File "/home//anaconda3/lib/python3.7/site-packages/torch/serialization.py", line 773, in _legacy_load
result = unpickler.load()
File "/home//anaconda3/lib/python3.7/site-packages/torch/serialization.py", line 729, in persistent_load
deserialized_objects[root_key] = restore_location(obj, location)
File "/home//anaconda3/lib/python3.7/site-packages/torch/serialization.py", line 178, in default_restore_location
result = fn(storage, location)
File "/home//anaconda3/lib/python3.7/site-packages/torch/serialization.py", line 158, in _cuda_deserialize
return storage_type(obj.size())
File "/home/*/anaconda3/lib/python3.7/site-packages/torch/cuda/init.py", line 433, in _lazy_new
return super(_CudaBase, cls).new(cls, *args, **kwargs)
RuntimeError: CUDA error: out of memory

[koyeongmin]

It looks like your GPU memory is lack. Could you check this point? In my case, I am using 2080ti, also you can control memory usage by changing batchsize.