search

paolorechia / learn-langchain

MIT License
275 stars 41 forks source link

Load different models command on windows #6

Closed unoriginalscreenname closed 1 year ago

unoriginalscreenname commented 1 year ago

Really love the repo here and the approach you're taking. I'm trying to get it to work and found one suggestion for your instructions. Your command to use a different model returns an error if you try and run it on windows:

export USE_13B_MODEL=true && export USE_4BIT=true && uvicorn servers.vicuna_server:app

The export and && isn't supported. I found that you can use the set command and format it like this:

set USE_13B_MODEL=true; set USE_4BIT=true;uvicorn servers.vicuna_server:app

paolorechia commented 1 year ago

That’s nice to know, I’m sure it will be helpful for others!

At some point I’ll reinstall this locally to add a proper README with all the required installation steps in Linux - if you would like, feel free to add something similar for Windows

paolorechia commented 1 year ago

Added these instructions to the README.md, thanks again.