Skip to content
This repository was archived by the owner on Feb 7, 2025. It is now read-only.

Conversation

@marksgraham
Copy link
Collaborator

@marksgraham marksgraham commented Jan 10, 2023

Fixes #160

@marksgraham marksgraham changed the title Pretrained DDPM [WIP] Pretrained DDPM Jan 10, 2023
@marksgraham marksgraham changed the title [WIP] Pretrained DDPM Pretrained DDPM Jan 11, 2023
@marksgraham marksgraham marked this pull request as ready for review January 11, 2023 21:02
@marksgraham
Copy link
Collaborator Author

Training with DDP on the KCL takes 7s/epoch using 2 GPUS, batch size 256 per GPU

@Warvito Warvito self-requested a review January 11, 2023 21:34
@marksgraham
Copy link
Collaborator Author

I'm attaching here the runai submission script, Dockerfile, and docker build script. They seemed too specific to our cluster to include in the actual PR
Archive.zip

Copy link
Collaborator

@Warvito Warvito left a comment

Choose a reason for hiding this comment

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

Looks great, Mark! Thank you for this great addition! I think just scaler in the evaluate functon might need some changes. But feel free to merge it after checking this review ^^

@marksgraham marksgraham merged commit d37a752 into main Jan 16, 2023
@Warvito Warvito deleted the 160-pretrained-ddpm branch February 11, 2023 15:01
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Provide pre-trained model option for tutorials

3 participants