All issues

Search results

Code release for generating synthetic data and training on larger scale data for CLIP?

Status: Open.

#10 In google-research/syn-rep-learn;

· EungyeupKim opened on Jan 1, 2025

How about the training GPU requirements?

Status: Open.

#9 In google-research/syn-rep-learn;

· qsunyuan opened on Sep 4, 2024

Question on MultiPosConLoss, how to change the loss to MoCo version?

Status: Open.

#8 In google-research/syn-rep-learn;

· yaoerqin opened on Aug 27, 2024

What is the specific method used in the paper 'Scaling law' to generate images using stable diffusion?

Status: Open.

#7 In google-research/syn-rep-learn;

· jizhejz opened on Jun 30, 2024

Have you evaluated CLIP and StableRep++ on imagenet zero-shot classification with laion50m subsets?

Status: Open.

#6 In google-research/syn-rep-learn;

· ytaek-oh opened on Mar 5, 2024

Prompt generation by SynCLR

Status: Open.

#5 In google-research/syn-rep-learn;

· yuntaodu opened on Feb 6, 2024

Linear probing duration

Status: Open.

#4 In google-research/syn-rep-learn;

· johanneskpp opened on Feb 2, 2024

Question on MultiPosConLoss and local_batch_size != self.last_local_batch_size

Status: Open.

#3 In google-research/syn-rep-learn;

· ArdalanM opened on Jan 12, 2024