Merge pull request #47 from chakki-works/fix/#46

Update validation to fix #46

seqeval/metrics/sequence_labeling.py

@@ -29,15 +29,15 @@ def get_entities(seq, suffix=False):
     """
 
     def _validate_chunk(chunk, suffix):
-        if chunk == 'O':
+        if chunk in ['O', 'B', 'I', 'E', 'S']:
             return
 
         if suffix:
-            if not (chunk.endswith('-B') or chunk.endswith('-I')):
+            if not (chunk.endswith('-B') or chunk.endswith('-I') or chunk.endswith('-E') or chunk.endswith('-S')):
                 raise ValueError('Invalid tag is found: {}'.format(chunk))
 
         else:
-            if not (chunk.startswith('B-') or chunk.startswith('I-')):
+            if not (chunk.startswith('B-') or chunk.startswith('I-') or chunk.startswith('E-') or chunk.startswith('S-')):
                 raise ValueError('Invalid tag is found: {}'.format(chunk))
 
     # for nested list

tests/test_metrics.py

@@ -49,6 +49,11 @@ def test_get_entities_with_unexpected_input_and_suffix_style(self):
         with self.assertRaises(Exception):
             get_entities(y_true)
 
+    def test_get_entities_with_only_IOB(self):
+        y_true = [['O', 'O', 'O', 'B', 'I', 'I', 'O'], ['B', 'I', 'O']]
+        entities = get_entities(y_true)
+        self.assertEqual(entities, [('_', 3, 5), ('_', 8, 9)])
+
     def test_performance_measure(self):
         y_true = [['O', 'O', 'O', 'B-MISC', 'I-MISC', 'O', 'B-ORG'], ['B-PER', 'I-PER', 'O', 'B-PER']]
         y_pred = [['O', 'O', 'B-MISC', 'I-MISC', 'I-MISC', 'O', 'O'], ['B-PER', 'I-PER', 'O', 'B-MISC']]