-
Notifications
You must be signed in to change notification settings - Fork 154
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
docs: update cifar_brevitas_training accuracy using representative calibration set #235
docs: update cifar_brevitas_training accuracy using representative calibration set #235
Conversation
| VGG FHE | 6 bits | 87.5\*\* | | ||
| VGG FHE (simulation\*) | 8 bits | 88.0 | | ||
| VGG FHE (simulation\*) | 7 bits | 87.2 | | ||
| VGG FHE (simulation\*) | 6 bits | 86.0 | |
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.
I removed the 5 bits because it gave ~60% accuracy, so not really relevant
| VGG FHE (simulation\*) | 8 bits | 88.0 | | ||
| VGG FHE (simulation\*) | 7 bits | 87.2 | | ||
| VGG FHE (simulation\*) | 6 bits | 86.0 | | ||
| VGG FHE | 6 bits | 86.0\*\* | |
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.
I'm currently executing the FHE inference on 10 samples to check for that
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.
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.
Thanks for the update. I think it's worth keeping a trace in this commit about why the accuracy changed since it's not in the same commit as the fix. Not sure where to add this.
EDIT: just understood your question actually, and Im also not sure how to do that (but not sure if it's really important) |
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.
Thanks
1efa947
to
7a3839b
Compare
It turns out there was not a bug, this is the new accuracy we should talk about. Basically, we were not properly calibrating before, but more details can be found in https://github.com/zama-ai/concrete-ml-internal/issues/3957
closes https://github.com/zama-ai/concrete-ml-internal/issues/3957