From 19f95f33ead4f6e217e4faeacbc860a7f10976fb Mon Sep 17 00:00:00 2001
From: George Lopatenko <81328772+Lopa10ko@users.noreply.github.com>
Date: Fri, 16 Aug 2024 17:54:45 +0300
Subject: [PATCH] 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
---
 .../data_operations/text_preprocessing.py                       | 2 +-
 other_requirements/extra.txt                                    | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)

diff --git a/fedot/core/operations/evaluation/operation_implementations/data_operations/text_preprocessing.py b/fedot/core/operations/evaluation/operation_implementations/data_operations/text_preprocessing.py
index 61a1da537d..929bfe5897 100644
--- a/fedot/core/operations/evaluation/operation_implementations/data_operations/text_preprocessing.py
+++ b/fedot/core/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
index 7aa4c932a9..cf2980ec4f 100644
--- 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