removed huggingface req

scoring-scripts/compute_seqeval.py

@@ -1,28 +1,49 @@
-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')
+from seqeval.metrics import classification_report, f1_score, precision_score, recall_score, accuracy_score
+from seqeval.scheme import IOB2
+import numpy as np
+def compute_seqeval_jsonl(references_jsonl, predictions_jsonl, ref_col='ner_tags', pred_col='pred_ner_tags'):
+    '''
+    Computes the seqeval scores between two datasets loaded from jsonl (list of dicts with same keys).
+    Sorts the datasets by 'unique_id' and verifies that the tokens match.
+    '''
+    # extract the tags and reverse the dict
+    ref_dict = {k:[e[k] for e in references_jsonl] for k in references_jsonl[0].keys()}
+    pred_dict = {k:[e[k] for e in predictions_jsonl] for k in predictions_jsonl[0].keys()}
+        
+    # sort by unique_id
+    ref_idx = np.argsort(ref_dict['unique_id'])
+    pred_idx = np.argsort(pred_dict['unique_id'])
+    ref_ner_tags = np.array(ref_dict[ref_col], dtype=object)[ref_idx]
+    pred_ner_tags = np.array(pred_dict[pred_col], dtype=object)[pred_idx]
+    ref_tokens = np.array(ref_dict['tokens'], dtype=object)[ref_idx]
+    pred_tokens = np.array(pred_dict['tokens'], dtype=object)[pred_idx]
 
     # 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,
-                       )
+    assert((ref_tokens==pred_tokens).all())
     
-    # change all values to regular (not numpy) floats (otherwise cannot be serialized to json)
-    seqeval_results = literal_eval(str(seqeval_results))
     
-    return(seqeval_results)
+    # get report
+    report = classification_report(y_true=ref_ner_tags, y_pred=pred_ner_tags, 
+                                   scheme=IOB2, output_dict=True,
+                                  )
+    
+    # extract values we care about
+    report.pop("macro avg")
+    report.pop("weighted avg")
+    overall_score = report.pop("micro avg")
+
+    seqeval_results = {
+        type_name: {
+            "precision": score["precision"],
+            "recall": score["recall"],
+            "f1": score["f1-score"],
+            "suport": score["support"],
+        }
+        for type_name, score in report.items()
+    }
+    seqeval_results["overall_precision"] = overall_score["precision"]
+    seqeval_results["overall_recall"] = overall_score["recall"]
+    seqeval_results["overall_f1"] = overall_score["f1-score"]
+    seqeval_results["overall_accuracy"] = accuracy_score(y_true=ref_ner_tags, y_pred=pred_ner_tags)    
+    
+    return(seqeval_results)