Skip to content
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.

Sign up for GitHub

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

Jump to bottom

Add notebook for fine-tuning YoloNAS with QAT #1638

Merged
merged 49 commits into from
Nov 23, 2023
Merged

Add notebook for fine-tuning YoloNAS with QAT #1638

merged 49 commits into from
Nov 23, 2023

Conversation

BloodAxe
Copy link
Collaborator

No description provided.

@shaydeci
Copy link
Collaborator

Few comments:

  • In section 1. you specify some info on multi_gpu (probably leftovers from the time that we passed it to Trainer ctor). Lets remove it or call setup_device somewhere just for demonstration (I think the rather makes more sense).
  • The helper method in 6. is basically Trainer.ptq(..).
  • I don't get the need for the call for Trainer.qat() in the last section. If we already called train on the model after replacing its layers with Q/DQ ones this is quite equivalent.
  • If we want to just demonstrate that qat can be done with 1 click and keep section 8. lets add some explanation about it and not leave it as "wrapping up"

@BloodAxe
Copy link
Collaborator Author

BloodAxe commented Nov 21, 2023 •
edited
Loading

Updated the notebook. @shaydeci can you please take a look?

shaydeci
shaydeci requested changes Nov 21, 2023
View reviewed changes
Copy link
Collaborator

@shaydeci shaydeci left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Notebook looks great.
There are some debugging leftovers I mentioned in the comments.
Also there's an empty code cell iat the bottom of the notebook.
Apart from those its LGTM on my side.

src/super_gradients/training/models/detection_models/yolo_nas/dfl_heads.py Outdated Show resolved Hide resolved
src/super_gradients/training/utils/callbacks/callbacks.py Outdated Show resolved Hide resolved
shaydeci
shaydeci approved these changes Nov 23, 2023
View reviewed changes
Copy link
Collaborator

@shaydeci shaydeci left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@BloodAxe BloodAxe merged commit 26d62e0 into master Nov 23, 2023
8 checks passed
@BloodAxe BloodAxe deleted the feature/SG-1214-update-yolo-nas-notebook branch November 23, 2023 08:38
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@shaydeci shaydeci shaydeci approved these changes

@ofrimasad ofrimasad Awaiting requested review from ofrimasad ofrimasad is a code owner

@Louis-Dupont Louis-Dupont Awaiting requested review from Louis-Dupont

Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.

None yet
2 participants
@BloodAxe @shaydeci