Changed regular expression to add support of the unicode characters p…

…rocessing and make code simple (#1411) Signed-off-by: igeni <kublin@it8.ru> Co-authored-by: intellinjun <105184542+intellinjun@users.noreply.github.com>
intel · Mar 25, 2024 · fd2516b · fd2516b
1 parent 3ae0cd0
commit fd2516b
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/...gface/pytorch/question-answering/deployment/squad/MLperf_example/datasets/tokenization.py b/...gface/pytorch/question-answering/deployment/squad/MLperf_example/datasets/tokenization.py
@@ -48,7 +48,7 @@ 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
 

diff --git a/...es/huggingface/pytorch/question-answering/deployment/squad/MLperf_example/tokenization.py b/...es/huggingface/pytorch/question-answering/deployment/squad/MLperf_example/tokenization.py
@@ -48,7 +48,7 @@ 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