scvi.dataloaders.AnnDataLoader#

class scvi.dataloaders.AnnDataLoader(adata_manager, shuffle=False, indices=None, batch_size=128, data_and_attributes=None, drop_last=False, iter_ndarray=False, **data_loader_kwargs)[source]#

DataLoader for loading tensors from AnnData objects.

Parameters:

  • adata_manager (AnnDataManager) – AnnDataManager object with a registered AnnData object.

  • shuffle – Whether the data should be shuffled

  • indices – The indices of the observations in the adata to load

  • batch_size (Optional[int]) – minibatch size to load each iteration

  • data_and_attributes (Optional[dict] (default: None)) – Dictionary with keys representing keys in data registry (adata_manager.data_registry) and value equal to desired numpy loading type (later made into torch tensor). If None, defaults to all registered data.

  • data_loader_kwargs – Keyword arguments for DataLoader

  • iter_ndarray (bool (default: False)) – Whether to iterate over numpy arrays instead of torch tensors

Attributes table#

multiprocessing_context

Methods table#

check_worker_number_rationality()

Attributes#

multiprocessing_context

AnnDataLoader.multiprocessing_context[source]#

dataset

AnnDataLoader.dataset: Dataset[T_co]#

batch_size

AnnDataLoader.batch_size: Optional[int]#

num_workers

AnnDataLoader.num_workers: int#

pin_memory

AnnDataLoader.pin_memory: bool#

drop_last

AnnDataLoader.drop_last: bool#

timeout

AnnDataLoader.timeout: float#

sampler

AnnDataLoader.sampler: Union[Sampler, Iterable]#

pin_memory_device

AnnDataLoader.pin_memory_device: str#

prefetch_factor

AnnDataLoader.prefetch_factor: int#

Methods#

check_worker_number_rationality

AnnDataLoader.check_worker_number_rationality()[source]#