minerva.data.datasets.context_dataset

Classes

ContextDataset

An abstract class representing a Dataset.

Module Contents

class minerva.data.datasets.context_dataset.ContextDataset(readers, transform=None)[source]

Bases: torch.utils.data.Dataset

An abstract class representing a Dataset.

All datasets that represent a map from keys to data samples should subclass it. All subclasses should overwrite __getitem__(), supporting fetching a data sample for a given key. Subclasses could also optionally overwrite __len__(), which is expected to return the size of the dataset by many Sampler implementations and the default options of DataLoader. Subclasses could also optionally implement __getitems__(), for speedup batched samples loading. This method accepts list of indices of samples of batch and returns list of samples.

Note

DataLoader by default constructs an index sampler that yields integral indices. To make it work with a map-style dataset with non-integral indices/keys, a custom sampler must be provided.

A PyTorch Dataset class for handling paired image and mask data with optional context transformations.

Parameters

readersTuple[_Reader, _Reader]

A tuple containing two reader objects. The first reader should provide images, and the second reader should provide corresponding masks. Both readers must support indexing and have the same length.

transformOptional[_Transform], default=None

An optional transformation function or callable that takes a tuple of (image, mask) and returns a transformed tuple of (image, mask). If None, no transformations are applied.

__getitem__(idx)[source]
Parameters:

idx (int)

Return type:

Tuple[numpy.ndarray, numpy.ndarray]

__len__()[source]
Return type:

int

readers
transform = None
Parameters: