Remove nltk punkt

marqo-ai · Nov 20, 2024 · 232e1ab · 232e1ab
1 parent c950504
commit 232e1ab
Showing 1 changed file with 1 addition and 5 deletions.
diff --git a/scripts/install_punkt_tokenizers.sh b/scripts/install_punkt_tokenizers.sh
@@ -1,13 +1,9 @@
 #!/bin/bash
-# This script is meant to be run at buildtime.
+# This script is meant to be run at build time.
 set -euo pipefail
 set -x
 
 mkdir -p /root/nltk_data/tokenizers
-curl https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/tokenizers/punkt.zip -o /root/nltk_data/tokenizers/punkt.zip
-unzip /root/nltk_data/tokenizers/punkt.zip -d /root/nltk_data/tokenizers/
-rm /root/nltk_data/tokenizers/punkt.zip
-
 curl https://raw.githubusercontent.com/nltk/nltk_data/gh-pages/packages/tokenizers/punkt_tab.zip -o /root/nltk_data/tokenizers/punkt_tab.zip
 unzip /root/nltk_data/tokenizers/punkt_tab.zip -d /root/nltk_data/tokenizers/
 rm /root/nltk_data/tokenizers/punkt_tab.zip