Issues

• Why normalize twice since the target encoder also normalize the feature at last？

  facebookresearch/ijepa#67 · Filter by author jianghaojun opened on Jul 17, 2024

• Question about dinov2 vs ijepa

  facebookresearch/ijepa#66 · Filter by author lanalex opened on Jul 3, 2024
  2

• How to create the visualizations on your README and Figure 6 of your paper?

  facebookresearch/ijepa#65 · Filter by author alanwang8 opened on Jun 22, 2024

• Why there is no **unscale_** when you use amp?

  facebookresearch/ijepa#64 · Filter by author jsrdcht opened on Jun 6, 2024

• Did you ever train with the ViTC?

  facebookresearch/ijepa#63 · Filter by author cvg25 opened on May 30, 2024

• Suggestions to train on low-end GPU

  facebookresearch/ijepa#62 · Filter by author ravioliasb opened on May 22, 2024
  5

• Why is there no class token?

  facebookresearch/ijepa#61 · Filter by author swarajnanda2021 opened on May 13, 2024
  4

• Bias in Multiblock Mask Collator

  facebookresearch/ijepa#60 · Filter by author joeagriffith opened on May 8, 2024

• Struggling to Train Downstream Classifier

  facebookresearch/ijepa#58 · Filter by author NolanGC opened on Apr 24, 2024
  5

• generative model decoder for target-encoder

  facebookresearch/ijepa#56 · Filter by author sls619 opened on Apr 1, 2024

• Training from scratch

  facebookresearch/ijepa#55 · Filter by author Ugenteraan opened on Mar 10, 2024
  13

• imagenet1k Huggingface extraction

  facebookresearch/ijepa#54 · Filter by author AasherH opened on Feb 4, 2024
  1

• Image resolution & folder structure for unsupervised pre-training

  facebookresearch/ijepa#53 · Filter by author rringham opened on Dec 11, 2023
  1

• Downstream task

  facebookresearch/ijepa#51 · Filter by author ankan8145 opened on Oct 29, 2023
  5

• How to load ijepa checkpoints?

  facebookresearch/ijepa#50 · Filter by author namrahrehman opened on Oct 10, 2023
  6

• Difficulty continue self supervised pre training on custom dataset

  facebookresearch/ijepa#49 · Filter by author ChristianEschen opened on Sep 16, 2023
  2

• An error was reported while training on multiple Gpus

  facebookresearch/ijepa#48 · Filter by author PeiranLiao opened on Aug 31, 2023
  1

• Struggling to replicate evaluation results

  facebookresearch/ijepa#47 · Filter by author finbarrtimbers opened on Aug 23, 2023
  14

• How to evaluate the performance of pre-trained model?

  facebookresearch/ijepa#46 · Filter by author xiangtaowong opened on Aug 23, 2023

• Is there any visual code?

  facebookresearch/ijepa#45 · Filter by author xiangtaowong opened on Aug 19, 2023

• I Just download the image1k dataset from huggingface, what should i do next to process the tar.gz file and train this model

  facebookresearch/ijepa#44 · Filter by author james2v opened on Aug 14, 2023
  3

• Error in interpolation of pos_embedd when using data of different dimension

  facebookresearch/ijepa#42 · Filter by author Pavinags opened on Jul 19, 2023

• config files about vit-small and vit-base

  facebookresearch/ijepa#39 · Filter by author stonewjf opened on Jul 17, 2023
  3

• is batch size per gpu important to re-implement the accuracy reported in the paper?

  facebookresearch/ijepa#35 · Filter by author Gus-Guo opened on Jul 3, 2023
  4

• Some questions about context and target

  facebookresearch/ijepa#32 · Filter by author lezhang7 opened on Jun 28, 2023
  1