Hi, I want to ask several questions related to load language model. Currently, I am trying to load Fasttext into Rasa by following this link, but I got this following error: “ModuleNotFoundError: No module named 'rasa_nlu_examples”. Any solution?
The other question is, any suggestion about language models that not resource hungry (like BERT and GPT) since I have been trying using BERT but I got an error said that OOM (Out of Memory).
Thanks in advance!
@prabowoyogawicaksana - you need to install the package
pip install "rasa_nlu_examples[fasttext]@https://github.com/RasaHQ/rasa-nlu-examples.git"
Thank you so much, but it didn’t worked in my case. I’m done with this problem by using this line python -m pip install "rasa_nlu_examples[fasttext] @ git+https://github.com/RasaHQ/rasa-nlu-examples.git" which derived from the complete documentation.
Anyway, thank you so much Sir
Hello Sir,
I am trying to use Fasttext vector via spacy following your medium blog, but I am unable to convert the vector file into spacy model as it shows the error: No module name ‘init-model’. “python -m spacy init --help” shows that it doesn’t have any command as init-model. Could you please help on how i can use the fasttext marathi words vector in rasa pipeline.
Hi, The blog i wrote was about 6 years old, i am pretty sure it is now outdated with the newer rasa versions.
Any particular reason you want to try fast text vectors?