Shitao commited on
Commit
cceec8c
1 Parent(s): 61ef075

Upload folder using huggingface_hub

Browse files
Files changed (1) hide show
  1. README.md +3 -3
README.md CHANGED
@@ -1793,9 +1793,9 @@ def get_new_queries(queries, query_max_len, examples_prefix, tokenizer):
1793
  return_tensors=None,
1794
  add_special_tokens=False
1795
  )
1796
- prefix_ids = tokenizer(examples_prefix)['input_ids']
1797
- suffix_ids = tokenizer('\n<response>')['input_ids']
1798
- new_max_length = (len(prefix_ids) + len(suffix_ids) + query_max_len) // 8 * 8 + 8
1799
  new_queries = tokenizer.batch_decode(inputs['input_ids'])
1800
  for i in range(len(new_queries)):
1801
  new_queries[i] = examples_prefix + new_queries[i] + '\n<response>'
 
1793
  return_tensors=None,
1794
  add_special_tokens=False
1795
  )
1796
+ prefix_ids = tokenizer(examples_prefix, add_special_tokens=False)['input_ids']
1797
+ suffix_ids = tokenizer('\n<response>', add_special_tokens=False)['input_ids']
1798
+ new_max_length = (len(prefix_ids) + len(suffix_ids) + query_max_len + 8) // 8 * 8 + 8
1799
  new_queries = tokenizer.batch_decode(inputs['input_ids'])
1800
  for i in range(len(new_queries)):
1801
  new_queries[i] = examples_prefix + new_queries[i] + '\n<response>'