scoring scripts

scoring-scripts/compute_MCC.py

@@ -0,0 +1,20 @@
+from sklearn.metrics import matthews_corrcoef
+import numpy as np
+def compute_MCC(references_dataset, predictions_dataset, ref_col='ner_tags', pred_col='pred_ner_tags'):
+    # computes the Matthews correlation coeff between two datasets
+    
+    # sort by id
+    references_dataset = references_dataset.sort('unique_id')
+    predictions_dataset = predictions_dataset.sort('unique_id')
+    
+    # check that tokens match
+    assert(references_dataset['tokens']==predictions_dataset['tokens'])
+    
+    # the lists have to be flattened
+    flat_ref_tags = np.concatenate(references_dataset[ref_col])
+    flat_pred_tags = np.concatenate(predictions_dataset[pred_col])
+    
+    mcc_score = matthews_corrcoef(y_true=flat_ref_tags,
+                                  y_pred=flat_pred_tags)
+    
+    return(mcc_score)

scoring-scripts/compute_seqeval.py

@@ -0,0 +1,28 @@
+from datasets import load_metric
+from ast import literal_eval
+def compute_seqeval(references_dataset, predictions_dataset, ref_col='ner_tags', pred_col='pred_ner_tags'):
+    # computes the seqeval scores
+
+    # sort by id
+    references_dataset = references_dataset.sort('unique_id')
+    predictions_dataset = predictions_dataset.sort('unique_id')
+    
+    # load the huggingface metric function
+    seqeval = load_metric('seqeval')
+
+    # check that tokens match
+    assert(references_dataset['tokens']==predictions_dataset['tokens'])
+
+    # ensure IOB2?
+
+    # compute scores
+    seqeval_results = seqeval.compute(predictions = predictions_dataset[pred_col], 
+                                      references = references_dataset[ref_col],
+                                      scheme = 'IOB2',
+                                      suffix = False,
+                       )
+    
+    # change all values to regular (not numpy) floats (otherwise cannot be serialized to json)
+    seqeval_results = literal_eval(str(seqeval_results))
+    
+    return(seqeval_results)