hotfix: pin nltk version to avoid downloading punkt_tab (#1321)

* fix: pin nltk to avoid downloading punkt_tab instead of punkt * fix: remove punkt_tab from nltk lookup list
aimclub · Aug 16, 2024 · 19f95f3 · 19f95f3
1 parent 39e960c
commit 19f95f3
Show file tree

Hide file tree

Showing 2 changed files with 2 additions and 2 deletions.
diff --git a/...ore/operations/evaluation/operation_implementations/data_operations/text_preprocessing.py b/...ore/operations/evaluation/operation_implementations/data_operations/text_preprocessing.py
@@ -65,7 +65,7 @@ def transform(self, input_data: InputData) -> OutputData:
 
  @staticmethod
  def _download_nltk_resources():
- for resource in ['punkt', 'punkt_tab']:
+ for resource in ['punkt']:
  try:
  nltk.data.find(f'tokenizers/{resource}')
  except LookupError:

diff --git a/other_requirements/extra.txt b/other_requirements/extra.txt
@@ -8,7 +8,7 @@ Pillow >= 8.2.0
 
 # Texts
 gensim==4.3.2
-nltk >= 3.5
+nltk==3.8.1
 
 # Misc
 protobuf~=3.19.0