I have created a chatbot using NLTK together with Python Flask that runs fine on my local Windows environment. When I deploy my code to Pythonanywhere I run into problems with the NLTK data. The error shows that it cannot find the wordnet.zip file:
Resource #033[93mwordnet#033[0m not found.
Please use the NLTK Downloader to obtain the resource:
**NO MATCH**
#033[31m>>> import nltk
>>> nltk.download('wordnet')
#033[0m
For more information see: https://www.nltk.org/data.html
**NO MATCH**
Attempted to load #033[93mcorpora/wordnet.zip/wordnet/#033[0m
**NO MATCH**
*Searched in:
- '**/home/vvephl2/nltk_data**'*
But the wordnet data is available at the given directory... see directory with wordnet.zip file
Until now I've tried:
Unzip wordnet.zip in dir corpora
Put wordnet.zip in dir nltk_data
Unzip wordnet.zip in dir nltk_data
re-install wordnet (nltk.download('wordnet'))
run tests using NLTK in Python interpreter; runs fine!!!
veryfied the NLTK configuration, looks fine
import nltk nltk.data.path ['/home/vvephl2/nltk_data', '/usr/local/nltk_data', '/usr/local/share/nltk_data', '/usr/local/lib/nltk_data', '/usr/share/nltk_data', '/usr/local/sh are/nltk_data', '/usr/lib/nltk_data', '/usr/local/lib/nltk_data']
What can I do to make the code on Pythonanywhere find the NLTK (wordnet) data??
Regards, Jos