I have a Dataset object that yields data generated from a pre-trained model that I load at the initialization of the dataset object. The model generating these data is placed on a separate GPU from the main model that will be optimized. Since the generation process is on a separate GPU, I would like to be able to generate more examples in parallel while the main model is performing an optimization step. I tried setting num_workers of the DataLoader object to a value greater than zero but I get the following error:
File "/usr/local/lib/python3.8/dist-packages/torch/utils/data/dataloader.py", line 279, in __iter__
return _MultiProcessingDataLoaderIter(self)
File "/usr/local/lib/python3.8/dist-packages/torch/utils/data/dataloader.py", line 719, in __init__
w.start()
File "/usr/lib/python3.8/multiprocessing/process.py", line 121, in start
self._popen = self._Popen(self)
File "/usr/lib/python3.8/multiprocessing/context.py", line 224, in _Popen
return _default_context.get_context().Process._Popen(process_obj)
File "/usr/lib/python3.8/multiprocessing/context.py", line 284, in _Popen
return Popen(process_obj)
File "/usr/lib/python3.8/multiprocessing/popen_spawn_posix.py", line 32, in __init__
super().__init__(process_obj)
File "/usr/lib/python3.8/multiprocessing/popen_fork.py", line 19, in __init__
self._launch(process_obj)
File "/usr/lib/python3.8/multiprocessing/popen_spawn_posix.py", line 47, in _launch
reduction.dump(process_obj, fp)
File "/usr/lib/python3.8/multiprocessing/reduction.py", line 60, in dump
ForkingPickler(file, protocol).dump(obj)
TypeError: cannot pickle 'module' object
Any suggestions on how to make this process faster in general?