metrics

Different Metric functions.

deeppavlov.metrics.accuracy.sets_accuracy(y_true: [<class 'list'>, <class 'numpy.ndarray'>], y_predicted: [<class 'list'>, <class 'numpy.ndarray'>]) float[source]

Calculate accuracy in terms of sets coincidence

Parameters
  • y_true – true values

  • y_predicted – predicted values

Returns

portion of samples with absolutely coincidental sets of predicted values

Alias:

sets_accuracy

deeppavlov.metrics.fmeasure.round_f1(y_true, y_predicted)[source]

Calculates F1 (binary) measure.

Parameters
  • y_true – list of true values

  • y_predicted – list of predicted values

Returns

F1 score

Alias:

f1

deeppavlov.metrics.fmeasure.round_f1_macro(y_true, y_predicted)[source]

Calculates F1 macro measure.

Parameters
  • y_true – list of true values

  • y_predicted – list of predicted values

Returns

F1 score

Alias:

f1_macro

deeppavlov.metrics.fmeasure.round_f1_weighted(y_true, y_predicted)[source]

Calculates F1 weighted measure.

Parameters
  • y_true – list of true values

  • y_predicted – list of predicted values

Returns

F1 score

Alias:

f1_weighted

deeppavlov.metrics.fmeasure.ner_f1(y_true, y_predicted)[source]

Calculates F1 measure for Named Entity Recognition task.

Parameters
  • y_true – list of true values

  • y_predicted – list of predicted values

Returns

F1 score

Alias:

ner_f1

deeppavlov.metrics.fmeasure.ner_token_f1(y_true, y_predicted, print_results=False)[source]

Calculates F1 measure for Named Entity Recognition task without taking into account BIO or BIOES markup.

Parameters
  • y_true – list of true values

  • y_predicted – list of predicted values

  • print_results – if True, then F1 score for each entity type is printed

Returns

F1 score

Alias:

ner_f1

deeppavlov.metrics.log_loss.sk_log_loss(y_true: Union[List[List[float]], List[List[int]], ndarray], y_predicted: Union[List[List[float]], List[List[int]], ndarray]) float[source]

Calculates log loss.

Parameters
  • y_true – list or array of true values

  • y_predicted – list or array of predicted values

Returns

Log loss

Alias:

log_loss

deeppavlov.metrics.roc_auc_score.roc_auc_score(y_true: Union[List[List[float]], List[List[int]], ndarray], y_pred: Union[List[List[float]], List[List[int]], ndarray]) float[source]

Compute Area Under the Curve (AUC) from prediction scores.

Parameters
  • y_true – true binary labels

  • y_pred – target scores, can either be probability estimates of the positive class

Returns

Area Under the Curve (AUC) from prediction scores

Alias:

roc_auc