We read every piece of feedback, and take your input very seriously.
To see all available qualifiers, see our documentation.
There was an error while loading. Please reload this page.
1 parent ec6bd00 commit 88094e0Copy full SHA for 88094e0
swift/llm/train/sft.py
@@ -232,13 +232,13 @@ def _prepare_callbacks(self):
232
self.callbacks = callbacks
233
234
def _stat_dataset(self, dataset: HfDataset):
235
- if not is_master():
236
- return
237
args = self.args
238
if isinstance(dataset, HfDataset):
239
dataset = GetLengthPreprocessor()(dataset, num_proc=args.dataset_num_proc)
240
length = dataset['length']
241
else:
+ if not is_master():
+ return
242
length = []
243
for row in tqdm(dataset, dynamic_ncols=True, desc='Get Length'):
244
length.append(max([len(row[k]) for k in row.keys() if k.endswith('input_ids')]))
0 commit comments