-
Notifications
You must be signed in to change notification settings - Fork 1.6k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Update huggingface.py #425
Conversation
Added quantization support using bitsandbytes
The latest updates on your projects. Learn more about Vercel for Git ↗︎
|
Codecov Report
@@ Coverage Diff @@
## main #425 +/- ##
==========================================
- Coverage 16.01% 15.97% -0.04%
==========================================
Files 59 59
Lines 2667 2673 +6
==========================================
Hits 427 427
- Misses 2240 2246 +6
📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more |
Perfect, thank you! Just make sure you import torch in the if q: section please |
Added import torch statement
I have done that just now, please check |
Congratulations @Archit-Kohli, you just earned a badge! Here it is: https://holopin.io/claim/clnd0foph78500flavc01bcao This badge can only be claimed by you, so make sure that your GitHub account is linked to your Holopin account. You can manage those preferences here: https://holopin.io/account. |
Added quantization support using bitsandbytes in application/llms/huggingface.py using simple q parameter for quantization support as in #409
What kind of change does this PR introduce? Feature Addition
Why was this change needed? Since the model is too large, quantization support is needed
Other information: Added bitsandbytes support for loading quantized model