Cannot load encoder on CPU with model trained on GPU

thedudeabides · June 27, 2019, 4:30pm

I trained a ULMFiT text model on a GPU, saved the model and the encoder. When I try

learn = text_classifier_learner(data_clas, AWD_LSTM, drop_mult=0.5)
learn.load_encoder()

on a CPU-only machine I get the first error message attached. When I follow the suggestion and instead call

learn.load_encoder(language_model_encoder_name, device=‘cpu’)

I get the same message (second attached message).

Any ideas?

Alex

RuntimeError Traceback (most recent call last)
in
----> 1 learn.load_encoder(language_model_encoder_name, device=‘cpu’)

~/anaconda3/lib/python3.7/site-packages/fastai/text/learner.py in load_encoder(self, name, device)
67 if device is None: device = self.data.device
68 if hasattr(encoder, ‘module’): encoder = encoder.module
—> 69 encoder.load_state_dict(torch.load(self.path/self.model_dir/f’{name}.pth’))
70 encoder.load_state_dict(torch.load(self.path/self.model_dir/f’{name}.pth’, map_location=device))
71 self.freeze()

~/anaconda3/lib/python3.7/site-packages/torch/serialization.py in load(f, map_location, pickle_module, **pickle_load_args)
385 f = f.open(‘rb’)
386 try:
–> 387 return _load(f, map_location, pickle_module, **pickle_load_args)
388 finally:
389 if new_fd:

~/anaconda3/lib/python3.7/site-packages/torch/serialization.py in _load(f, map_location, pickle_module, **pickle_load_args)
572 unpickler = pickle_module.Unpickler(f, **pickle_load_args)
573 unpickler.persistent_load = persistent_load
–> 574 result = unpickler.load()
575
576 deserialized_storage_keys = pickle_module.load(f, **pickle_load_args)

~/anaconda3/lib/python3.7/site-packages/torch/serialization.py in persistent_load(saved_id)
535 obj = data_type(size)
536 obj._torch_load_uninitialized = True
–> 537 deserialized_objects[root_key] = restore_location(obj, location)
538 storage = deserialized_objects[root_key]
539 if view_metadata is not None:

~/anaconda3/lib/python3.7/site-packages/torch/serialization.py in default_restore_location(storage, location)
117 def default_restore_location(storage, location):
118 for _, _, fn in _package_registry:
–> 119 result = fn(storage, location)
120 if result is not None:
121 return result

~/anaconda3/lib/python3.7/site-packages/torch/serialization.py in _cuda_deserialize(obj, location)
93 def _cuda_deserialize(obj, location):
94 if location.startswith(‘cuda’):
—> 95 device = validate_cuda_device(location)
96 if getattr(obj, “_torch_load_uninitialized”, False):
97 storage_type = getattr(torch.cuda, type(obj).name)

~/anaconda3/lib/python3.7/site-packages/torch/serialization.py in validate_cuda_device(location)
77
78 if not torch.cuda.is_available():
—> 79 raise RuntimeError('Attempting to deserialize object on a CUDA ’
80 'device but torch.cuda.is_available() is False. ’
81 'If you are running on a CPU-only machine, ’

RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=‘cpu’ to map your storages to the CPU.

With ‘cpu’ specified as device: