sfaira.estimators.EstimatorKerasCelltype.evaluate_any

EstimatorKerasCelltype.evaluate_any(idx, batch_size: int = 128, max_steps: int = inf, weighted: bool = False)

Evaluate the custom model on any local data.

Defaults to run on full data if idx is None.

Parameters

• idx – Indices of observations to evaluate on. Evaluates on all observations if None.

• batch_size – Batch size for evaluation.

• max_steps – Maximum steps before evaluation round is considered complete.

• weighted – Whether to use class weights in evaluation.

Returns

Dictionary of metric names and values.