I have Rasa installed. So I was wondering if I had to explicitly install spaCy to be able to use it in the pipeline or not? Can I somehow just fit the spaCyNLU, spaCy tokenizer into the pipeline, without installing it?
If installing is necessary, can someone please suggest a guide that teaches how to install it, provided that we have rasa installed.
You need to be able to use SpaCy in order to use it. Execute these commands:
pip install rasa[spacy]
python -m spacy download en_core_web_md
If you need to link en to en_core_web_md, the command will output what you need to do.
Hello, I am trying this command in pycharm pip install rasa[spacy] and last 15 hours the installation is going on but still not complete. It is at… Collecting urllib3<1.27,>=1.25.4 Using cached urllib3-1.26.7-py2.py3-none-any.whl (138 kB) Using cached urllib3-1.26.6-py2.py3-none-any.whl (138 kB)
and still not sure how long it will take. can somebody help me to get it installed quicker pls.
@nlu Try to add --use-deprecated=legacy-resolver in your pip command
thank you so much for quick reply. It did install in less than an hour. There are some red warnings that I will look into . But thank you very much.
@nlu even try downgrade the pip version and install again.
@nik202 , Thank you for your reply. Will do that as well.
Thanks .
No need to downgrade if you already used –use-deprecated=legacy-resolver, these are two alternative
solutions and not complementary solutions. No need to do one if you did the other.
If anything, I would suggest upgrading PIP since this is fixed in the newer versions.
To fix those, you can run the command without –use-deprecated=legacy-resolver again 