cascade.data.configure_dataset

cascade.data.configure_dataset(adata, use_regime=None, use_covariate=None, use_size=None, use_weight=None, use_layer=None)[source]

Configure dataset for model training and inferences

Parameters:

  • adata (AnnData) – Input dataset

  • use_regime (str | None) – Key in layers containing intervention regime encoded with encode_regime()

  • use_covariate (str | None) – Key in obsm containing covariates

  • use_size (str | None) – Key in obs containing cell sizes

  • use_weight (str | None) – Key in obs containing cell weights

  • use_layer (str | None) – Key in layers containing data to be used for training and inferences, instead of X

Return type:

None

Note

When called multiple times, individual configuration items will be overwritten if and only if new values are not None.