prefetch_factor

[zhuang-maowei]

fastNLP\core\dataloaders\torch_dataloader\fdl.py in init(self, dataset, batch_size, shuffle, sampler, batch_sampler, num_workers, collate_fn, pin_memory, drop_last, timeout, worker_init_fn, multiprocessing_context, generator, prefetch_factor, persistent_workers, **kwargs)
150 persistent_workers=persistent_workers)
...
--> 245 raise ValueError('prefetch_factor option could only be specified in multiprocessing.'
246 'let num_workers > 0 to enable multiprocessing, otherwise set prefetch_factor to None.')
247 elif num_workers > 0 and prefetch_factor is None:

ValueError: prefetch_factor option could only be specified in multiprocessing.let num_workers > 0 to enable multiprocessing, otherwise set prefetch_factor to None.

[diyuhuo9977]

have same problem, set prefetch_factor param value to None can fix this error
fastNLP==1.0.1
fasthan==2.0
path:
FastModel.py#__preprocess_sentence

dataloader = TorchDataLoader(dataset, **prefetch_factor=None,**
                                     batch_sampler=BucketedBatchSampler(
                                         dataset=dataset,
                                         batch_size=1,
                                         length='seq_len'))