class documentation

class BatchCompose(A.Compose):

Constructor: BatchCompose(transforms, **kwargs)

View In Hierarchy

Compose batch-aware Albumentations transforms.

Method __call__ Apply the composed transforms to a batch.
Method __init__ Compose batch transforms.
Instance Variable batch_size Product of nested transform batch sizes, ibi.
Instance Variable transforms Batch transformations in composition order.
Static Method _make_contiguous Undocumented
@override
def __call__(self, data_batch: list[dict[str, np.ndarray]]) -> dict[str, np.ndarray]:

Apply the composed transforms to a batch.

Parameters
data_batch:list[dict[str, np.ndarray]]Batch of Albumentations data dictionaries. Its length must equal batch_size.
Returns
dict[str, np.ndarray]Single transformed data dictionary.
Raises
ValueErrorIf len(data_batch) does not match batch_size.
def __init__(self, transforms: TransformsSeqType, **kwargs):

Compose batch transforms.

Parameters
transforms:TransformsSeqTypeTransformations to compose.
**kwargsAdditional arguments passed to A.Compose.
batch_size: int =

Product of nested transform batch sizes, ibi.

transforms: list[BatchTransform] =

Batch transformations in composition order.

@staticmethod
def _make_contiguous(data: dict[str, np.ndarray]) -> dict[str, np.ndarray]:

Undocumented