Search results

Loss increasing

Status: Open.

#50 In facebookresearch/simsiam;

· karthikpullalarevu opened on Aug 9, 2023

Could not find any class folder

Status: Open.

#49 In facebookresearch/simsiam;

· chandagrover opened on Apr 16, 2023

Why are training and testing so slow

Status: Open.

#48 In facebookresearch/simsiam;

· InstantWindy opened on Mar 6, 2023

which vector is used in experiment (the vector atfer predictor or after projection)?

Status: Open.

#47 In facebookresearch/simsiam;

· aruba01 opened on Feb 26, 2023

The value of loss function nn.CosineSimilarity is negative

Status: Open.

#44 In facebookresearch/simsiam;

· zhang1hongliang opened on Nov 3, 2022

Pre-trained weights for SimCLR, MoCov2, BYOL, SwaV

Status: Open.

#43 In facebookresearch/simsiam;

· ninatu opened on Oct 27, 2022

About the shape of image[0] and image[1]

Status: Open.

#42 In facebookresearch/simsiam;

· Richard-Lu-badbird opened on Oct 16, 2022

Loss collapse during training

Status: Open.

#40 In facebookresearch/simsiam;

· aisagarw opened on Oct 5, 2022

About the projection and prediction head dimension configs

Status: Open.

#39 In facebookresearch/simsiam;

· Dylan-H-Wang opened on Aug 30, 2022

Slow convergence with SGD linear evaluation

Status: Open.

#37 In facebookresearch/simsiam;

· gergopool opened on May 10, 2022

checkpoint for epoch 800

Status: Open.

#36 In facebookresearch/simsiam;

· elias-ramzi opened on May 4, 2022

larger batch size with linear scale does not work

Status: Open.

#35 In facebookresearch/simsiam;

· liming-ai opened on May 4, 2022