metrics_table#

er_evaluation.metrics_table(predictions, references, metrics={'B-Cubed F1': <function b_cubed_f>, 'B-Cubed Precision': <function b_cubed_precision>, 'B-Cubed Recall': <function b_cubed_recall>, 'Cluster F1': <function cluster_f>, 'Cluster Precision': <function cluster_precision>, 'Cluster Recall': <function cluster_recall>, 'Pairwise F1': <function pairwise_f>, 'Pairwise Precision': <function pairwise_precision>, 'Pairwise Recall': <function pairwise_recall>})[source]#

Apply a set of metrics to all combinations of prediction and reference membership vectors.

Parameters:

  • predictions (Dict) – Dictionary of membership vectors.

  • references (Dict) – Dictionary of membership vectors.

  • metrics (Dict) – Dictionary of metrics to apply to the prediction and reference pairs.

Returns:

Dataframe with columns “prediction”, “reference”, “metric”, and “value”, containing the value of the given metric applied to the corresponding prediction and reference membership vector.

Return type:

DataFrame

Examples

>>> predictions = {"prediction_1": pd.Series(index=[1,2,3,4,5,6,7,8], data=[1,1,2,3,2,4,4,4])}
>>> references = {"reference_1": pd.Series(index=[1,2,3,4,5,6,7,8], data=["c1", "c1", "c1", "c2", "c2", "c3", "c3", "c4"])}
>>> metrics = {"precision": pairwise_precision, "recall": pairwise_recall}
>>> metrics_table(predictions, references, metrics) 
    prediction      reference   metric      value
0       prediction_1    reference_1     precision       0.4
1       prediction_1    reference_1     recall      0.4