nntoolbox.utils.data module

class nntoolbox.utils.data.SupervisedDataset(inputs: numpy.ndarray, labels: numpy.ndarray, device=device(type='cpu'), transform=None)[source]

Bases: Generic[torch.utils.data.dataset.T_co]

classmethod from_csv(path: str, label_name: str, data_fields: Optional[List[str]] = None, device=device(type='cpu'))[source]

Create a supervised dataset from csv file

prepare_arr(tensor: torch.Tensor, dtype)[source]
nntoolbox.utils.data.get_first_batch(data: torch.utils.data.dataloader.DataLoader, callbacks: Optional[Iterable[Callback]] = None)[source]

Get the first batch from dataloader

Parameters

  • data – the dataloader

  • callbacks – the list of callbacks to applied to data

nntoolbox.utils.data.grab_next_batch(data: Union[torch.utils.data.dataloader.DataLoader, torchtext.data.iterator.Iterator])[source]

Grab the next batch from dataloader