Evaluator

Evaluator for testing.

class mindnlp.engine.evaluator.Evaluator(network, eval_dataset=None, metrics=None, callbacks=None, jit=False)[source]

Bases: object

Evaluator to test the model.

Parameters:

  • network (Cell) – A network for evaluating.

  • eval_dataset (Dataset) – A evaluating dataset iterator.

  • batc_size (int) – numbers of samples in each batch.

  • metrcis (Optional[list[Metric], Metric]) – List of metric objects which should be used while evaluating. Default:None.

  • callbacks (Optional[list[Callback], Callback]) – List of callback objects which should be executed while training. Default: None.

  • jit (bool) – Whether use Just-In-Time compile.

clear_metrics()[source]

Clear metrics values.

run(tgt_columns=None)[source]

Evaluating function entry.

Parameters:

tgt_columns (Optional[list[str], str]) – Target label column names for loss function.