-
Notifications
You must be signed in to change notification settings - Fork 3
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
additional fixes for HFQuantizer compatibility #136
Conversation
@@ -205,6 +205,10 @@ def parse_quantization_config( | |||
if hasattr(compression_config, QUANTIZATION_CONFIG_NAME): | |||
# for loaded HFQuantizer config | |||
return getattr(compression_config, QUANTIZATION_CONFIG_NAME) | |||
elif isinstance(compression_config, dict) and ( |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
already merged
@kylesayrs @dsikka is this diff still needed and in progress given the landing of HF quantizer support? |
parameter.data = new_param_data.to(device).to(dtype) | ||
try: | ||
parameter.data = new_param_data.to(device).to(dtype) | ||
except RuntimeError: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
This is better handled by #193
All the changes here have already been made or have open PRs to fix |
handling of assigning values to meta parameters on load and deserializing quant configs