All issues

Search results

Performance drops on 256x256 images

Status: Open.

#8In JoshVarty/SelfSupervisedLearning;

· JoshVarty opened on Feb 7, 2020

Consider model weight averaging after pretext task?

Status: Open.

#7In JoshVarty/SelfSupervisedLearning;

· JoshVarty opened on Feb 4, 2020

Make sure we're normalizing with imagenet_stats

Status: Open.

#6In JoshVarty/SelfSupervisedLearning;

· JoshVarty opened on Jan 29, 2020

Test effects of more and fewer patches on downstream performance

Status: Open.

#5In JoshVarty/SelfSupervisedLearning;

· JoshVarty opened on Jan 29, 2020

Consider removing some of the UNet connections

Status: Open.

#4In JoshVarty/SelfSupervisedLearning;

· JoshVarty opened on Jan 23, 2020

Consider only calculating loss over patches

Status: Open.

#3In JoshVarty/SelfSupervisedLearning;

· JoshVarty opened on Jan 23, 2020