changed tokenization
Browse files- ref_seg_ger.py +1 -1
ref_seg_ger.py
CHANGED
|
@@ -229,7 +229,7 @@ class RefSeg(datasets.GeneratorBasedBuilder):
|
|
| 229 |
|
| 230 |
# tokenized_input = row['token'].split(' ')
|
| 231 |
tkn = self.TOKENIZER.pre_tokenize_str(row['token'])
|
| 232 |
-
if row['label'] == '
|
| 233 |
row['label'] = 'other'
|
| 234 |
if not tkn:
|
| 235 |
continue
|
|
|
|
| 229 |
|
| 230 |
# tokenized_input = row['token'].split(' ')
|
| 231 |
tkn = self.TOKENIZER.pre_tokenize_str(row['token'])
|
| 232 |
+
if row['label'] == 'identifier':
|
| 233 |
row['label'] = 'other'
|
| 234 |
if not tkn:
|
| 235 |
continue
|