Coding tutorial: RAG with LangChain and Llama3

[@ygfster]

Nice video and notebook, thanks for sharing!

[@akilkhatri1089]

Thank you!!

[@kylewilliams4721]

So this isn't a local Llama3 model, does huggingface store api request data?

[@yanaitalk]

This is a local model, downloaded from huggingface.

[@kylewilliams4721]

@@yanaitalk Just to be clear, you downloaded the llama3 model on your hard drive?

[@yanaitalk]

@@kylewilliams4721 yes, in the notebook, it is downloaded to local drive in colab.

[@AhsenWaheed-230]

​@@yanaitalk
I am getting this error : No module named 'transformers.models.cohere.configuration_cohere' at line [llama3 = AutoModelForCausalLM.from_pretrained(..)]....somewhere on internet i found that I need to update the transformer version and I changed the version in requirements.txt and used the latest version.But I am still getting this error. Can anyone help with that plz.
Also, I just download the whole model from huggingface and uploaded to my drive. Can I use it directly instead of downloading it using the above line where I am getting the error.

[@amitsingha1637]

Thanks

[@ashwinkumar5223]

Superb.. Thanks