Search results

L2 regulariser and SAM

Status: Open.

#24 In google-research/sam;

· konstantinos-p opened on Aug 20, 2023

Something wrong before the second loss backward

Status: Open.

#17 In google-research/sam;

· KeranLi opened on Aug 4, 2023

Implementation of the gradient approximation for the SAM objective (eq3)

Status: Open.

#14 In google-research/sam;

· zhipeng-wei opened on Nov 20, 2021

Implementation details of grad_norm computation

Status: Open.

#13 In google-research/sam;

· shuo-ouyang opened on Aug 9, 2021

Unable to reproduce results with CIFAR-10 dataset on efficientnet-l2-475

Status: Open.

#12 In google-research/sam;

· MiguelCosta94 opened on Jul 26, 2021

Error reproducing results with efficientnet on SVHN dataset, the size of input is more then 224, but image is 32x32

Status: Open.

#10 In google-research/sam;

· MotiBaadror opened on Jul 22, 2021

Error running examples

Status: Open.

#9 In google-research/sam;

· le000043 opened on May 24, 2021

Verification on Training Details

Status: Open.

#8 In google-research/sam;

· lthilnklover opened on May 19, 2021

Reproducing WRN-28-10 (SAM) for SVHN dataset

Status: Open.

#7 In google-research/sam;

· akshaydp1995 opened on Apr 5, 2021

Reproducibility Issues

Status: Open.

#6 In google-research/sam;

· devansh20la opened on Mar 29, 2021

Regarding mentioning a TF2 implementation

Status: Open.

#5 In google-research/sam;

· sayakpaul opened on Feb 25, 2021

Accelerated training with floating point fp16

Status: Open.

#3 In google-research/sam;

· milliema opened on Jan 10, 2021