Pytorch AssertionError: Torch not compiled with CUDA enabled

Question

I am trying to run code from this repo. I have disabled cuda by changing lines 39/40 in main.py from

parser.add_argument('--type', default='torch.cuda.FloatTensor', help='type of tensor - e.g torch.cuda.HalfTensor')

to

parser.add_argument('--type', default='torch.FloatTensor', help='type of tensor - e.g torch.HalfTensor')

Despite this, running the code gives me the following exception:

Traceback (most recent call last):
  File "main.py", line 190, in <module>
    main()
  File "main.py", line 178, in main
    model, train_data, training=True, optimizer=optimizer)
  File "main.py", line 135, in forward
    for i, (imgs, (captions, lengths)) in enumerate(data):
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 201, in __next__
    return self._process_next_batch(batch)
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 221, in _process_next_batch
    raise batch.exc_type(batch.exc_msg)
AssertionError: Traceback (most recent call last):
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 62, in _pin_memory_loop
    batch = pin_memory_batch(batch)
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 123, in pin_memory_batch
    return [pin_memory_batch(sample) for sample in batch]
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 123, in <listcomp>
    return [pin_memory_batch(sample) for sample in batch]
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/utils/data/dataloader.py", line 117, in pin_memory_batch
    return batch.pin_memory()
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/tensor.py", line 82, in pin_memory
    return type(self)().set_(storage.pin_memory()).view_as(self)
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/storage.py", line 83, in pin_memory
    allocator = torch.cuda._host_allocator()
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/cuda/__init__.py", line 220, in _host_allocator
    _lazy_init()
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/cuda/__init__.py", line 84, in _lazy_init
    _check_driver()
  File "/Users/lakshay/anaconda/lib/python3.6/site-packages/torch/cuda/__init__.py", line 51, in _check_driver
    raise AssertionError("Torch not compiled with CUDA enabled")
AssertionError: Torch not compiled with CUDA enabled

Spent some time looking through the issues in the Pytorch github, to no avail. Help, please?

Answer 1

If you look into the data.py file, you can see the function:

def get_iterator(data, batch_size=32, max_length=30, shuffle=True, num_workers=4, pin_memory=True):
    cap, vocab = data
    return torch.utils.data.DataLoader(
        cap,
        batch_size=batch_size, shuffle=shuffle,
        collate_fn=create_batches(vocab, max_length),
        num_workers=num_workers, pin_memory=pin_memory)

which is called twice in main.py file to get an iterator for the train and dev data. If you see the DataLoader class in pytorch, there is a parameter called:

pin_memory (bool, optional) – If True, the data loader will copy tensors into CUDA pinned memory before returning them.

which is by default True in the get_iterator function. And as a result you are getting this error. You can simply pass the pin_memory param value as False when you are calling get_iterator function as follows.

train_data = get_iterator(get_coco_data(vocab, train=True),
                          batch_size=args.batch_size,
                          ...,
                          ...,
                          ...,
                          pin_memory=False)