-
Notifications
You must be signed in to change notification settings - Fork 188
SD tutorial and benchmarks #90
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
Conversation
jeffra
commented
Nov 1, 2022
- SD benchmark scripts
- SD tutorial showing how to deploy and evaluate a deployment
- Add MII package version that can be checked at runtime
- Fix GPRC max message size issue that I ran into when running multi-batch SD
- Added optional mii config to skip model validation checks (this currently takes >15sec and is not always needed)
mrwyattii
left a comment
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.
LGTM, just a few minor suggestions to tutorial
A bit of restructuring
Co-authored-by: Michael Wyatt <michaelwyatt@microsoft.com>
Co-authored-by: Michael Wyatt <michaelwyatt@microsoft.com>
|
I'd recommend you mention how many steps are taken for the benchmark comparison. Same with image dimensions. Both really impact throughput. Maybe try compare variations like 512x512 vs 1024x1024. I didn't immediately find it so apologies if I missed it too. |
|
Thank you @Suhail for the feedback, I’ll update the tutorial before merging to add this info. One thing for anyone reading this PR, we have pending changes coming later today on the DeepSpeed side related further performance updates and bug fixes. We’ll merge this PR once those our live. |