improvement of regexp (including support of unicode)

2025-07-23 03:33:05 +00:00 · 2024-03-21 22:05:04 +03:00 · 2024-03-21 22:05:04 +03:00 · 105d9745b1
commit 105d9745b1
parent 848a574c26
1 changed files with 1 additions and 1 deletions
--- a/examples/tutorial/sequence_parallel/data/tokenizer/bert_tokenization.py
+++ b/examples/tutorial/sequence_parallel/data/tokenizer/bert_tokenization.py
@ -35,7 +35,7 @@ def validate_case_matches_checkpoint(do_lower_case, init_checkpoint):
    if not init_checkpoint:
        return

-    m = re.match("^.*?([A-Za-z0-9_-]+)/bert_model.ckpt", init_checkpoint)
+    m = re.match("^.*?([\w-]+)/bert_model.ckpt", init_checkpoint)
    if m is None:
        return