cellflow.data.TrainingData¶

class cellflow.data.TrainingData(cell_data, split_covariates_mask, split_idx_to_covariates, perturbation_covariates_mask, perturbation_idx_to_covariates, perturbation_idx_to_id, condition_data, control_to_perturbation, max_combination_length, null_value, data_manager)[source]¶

Training data.

Parameters:

cell_data (ndarray) – The representation of cell data, e.g. PCA of gene expression data.
split_covariates_mask (ndarray) – Mask of the split covariates.
split_idx_to_covariates (dict[int, tuple[Any, ...]]) – Dictionary explaining values in split_covariates_mask.
perturbation_covariates_mask (ndarray) – Mask of the perturbation covariates.
perturbation_idx_to_covariates (dict[int, tuple[str, ...]]) – Dictionary explaining values in perturbation_covariates_mask.
condition_data (dict[str, ndarray]) – Dictionary with embeddings for conditions.
control_to_perturbation (dict[int, ndarray]) – Mapping from control index to target distribution indices.
max_combination_length (int) – Maximum number of covariates in a combination.
data_manager (Any) – The data manager
perturbation_idx_to_id (dict[int, Any])
null_value (Any)

Methods

Attributes

`n_controls`	Returns the number of control covariate values.
`n_perturbation_covariates`	Returns the number of perturbation covariates.
`n_perturbations`	Returns the number of perturbation covariate combinations.
`cell_data`
`split_covariates_mask`
`split_idx_to_covariates`
`perturbation_covariates_mask`
`perturbation_idx_to_covariates`
`perturbation_idx_to_id`
`condition_data`
`control_to_perturbation`
`max_combination_length`
`null_value`
`data_manager`