supawichwac commited on
Commit
12b47c1
1 Parent(s): 56a94af

Saving train state of step 5

Browse files
distil-whisper/events.out.tfevents.1715073503.server02.1431090.0 ADDED
@@ -0,0 +1,3 @@
 
 
 
 
1
+ version https://git-lfs.github.com/spec/v1
2
+ oid sha256:ce53bdcb28f245a46122283081f5c6c2f3e33cca3de217f4fa873fc9b8143fec
3
+ size 88
run_distillation.py CHANGED
@@ -874,7 +874,7 @@ def main():
874
  raw_datasets["eval"] = load_dataset(
875
  dataset_dict["name"],
876
  dataset_dict["config"],
877
- split=dataset_dict["split"],
878
  cache_dir=data_args.dataset_cache_dir,
879
  token=model_args.token,
880
  streaming=data_args.streaming,
@@ -1595,9 +1595,8 @@ def main():
1595
  eval_preds = []
1596
  eval_labels = []
1597
  eval_start = time.time()
1598
- print(f" vectorized_datasets[eval_split] : {len(vectorized_datasets[eval_split])}")
1599
-
1600
- validation_dataloader = DataLoader(
1601
  vectorized_datasets[eval_split],
1602
  collate_fn=data_collator,
1603
  batch_size=per_device_eval_batch_size,
 
874
  raw_datasets["eval"] = load_dataset(
875
  dataset_dict["name"],
876
  dataset_dict["config"],
877
+ split=dataset_dict["split"],
878
  cache_dir=data_args.dataset_cache_dir,
879
  token=model_args.token,
880
  streaming=data_args.streaming,
 
1595
  eval_preds = []
1596
  eval_labels = []
1597
  eval_start = time.time()
1598
+
1599
+ F = DataLoader(
 
1600
  vectorized_datasets[eval_split],
1601
  collate_fn=data_collator,
1602
  batch_size=per_device_eval_batch_size,