From 98c4c06cf75d4c5aaef5fb476e1402b6d593f349 Mon Sep 17 00:00:00 2001 From: =?UTF-8?q?=D0=90=D0=BB=D0=B5=D0=BA=D1=81=D0=B5=D0=B9=20=D0=90=D0=BD?= =?UTF-8?q?=D1=82=D0=BE=D0=BD=D0=BE=D0=B2?= Date: Thu, 12 Mar 2026 15:44:36 +0300 Subject: [PATCH] fix punkt_tab (cherry picked from commit a819afcffd0cb5ec2a93606edc18f0d16076ebcc) --- smart_kit/text_preprocessing/local_text_normalizer.py | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/smart_kit/text_preprocessing/local_text_normalizer.py b/smart_kit/text_preprocessing/local_text_normalizer.py index a1df191e..399f1e4c 100644 --- a/smart_kit/text_preprocessing/local_text_normalizer.py +++ b/smart_kit/text_preprocessing/local_text_normalizer.py @@ -40,7 +40,7 @@ def get_token_list(self, text): def __load_everything(self): try: - nltk.data.find("tokenizers/punkt") + nltk.data.find("tokenizers/punkt_tab") except LookupError: nltk.download("punkt_tab")