kindred.evaluate¶
-
kindred.
evaluate
(goldCorpus, testCorpus, metric='f1score', display=False)[source]¶ Compares the gold corpus with the test corpus and calculate appropriate metrics.
Parameters: - goldCorpus (kindred.Corpus) – The gold standard set of data
- testCorpus (kindred.Corpus) – The test set for comparison
- metric (str) – Which metric to use (precision/recall/f1score). ‘all’ will provide all three as a tuple
- display (bool) – Whether to print (to stdout) specific statistics for each relation type
Returns: The value of the corresponding metric (or metrics)
Return type: float (or tuple of floats)