torch_mist.utils.data

Submodules

Package Contents

Classes

SameAttributeDataLoader

SameAttributeSampler

SampleDataset

All the operations on a read-only sequence.

CSVDataset

Functions

sample_same_attributes(→ torch.utils.data.DataLoader)

class torch_mist.utils.data.SameAttributeDataLoader(dataset: torch.utils.data.Dataset, attributes: torch.Tensor | numpy.ndarray, batch_size: int, neg_samples: int, **kwargs)

Bases: torch.utils.data.DataLoader

torch_mist.utils.data.sample_same_attributes(dataloader: torch.utils.data.DataLoader, attributes: torch.Tensor | numpy.ndarray, neg_samples: int) → torch.utils.data.DataLoader
class torch_mist.utils.data.SameAttributeSampler(batch_size: int, neg_samples: int, attributes: torch.Tensor | numpy.ndarray)

Bases: torch.utils.data.Sampler

__iter__()
__len__()
class torch_mist.utils.data.SampleDataset(samples: Dict[str, torch.Tensor])

Bases: Sequence, torch.utils.data.Dataset

All the operations on a read-only sequence.

Concrete subclasses must override __new__ or __init__, __getitem__, and __len__.

__getitem__(item)
__len__()
class torch_mist.utils.data.CSVDataset(filepath: str, remove_nan_rows: bool = True, rename_columns: Dict[str, str] | None = None, **kwargs)

Bases: torch.utils.data.Dataset

__len__()
__getitem__(index)