scglue.models.data.AnnDataset

class scglue.models.data.AnnDataset(adatas, data_configs, mode='train', getitem_size=1)[source]

Bases: Dataset

Dataset for anndata.AnnData objects with partial pairing support.

Parameters
  • *adatas – An arbitrary number of configured anndata.AnnData objects

  • data_configs (typing.List[typing.Mapping[str, typing.Any]]) – Data configurations, one per dataset

  • mode (str) – Data mode, must be one of {"train", "eval"}

  • getitem_size (int) – Unitary fetch size for each __getitem__ call

Methods

accept_shuffle

Accept shuffling result

propose_shuffle

Propose shuffling using a given random seed

random_split

Randomly split the dataset into multiple subdatasets according to given fractions.

Attributes

adatas

Internal AnnData objects

data_configs

Data configuration for each dataset

logger