Skip to content

Issues: facebookresearch/dinov2

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

231 Open 192 Closed
231 Open 192 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Issues list

height and width order mixed up
#504 opened Mar 4, 2025 by J4Q8
Reproducing Depth results using Monocular Depth Estimation Toolbox
#503 opened Mar 3, 2025 by davnords
1
deduplication
#502 opened Feb 27, 2025 by SebastianJanampa
Unauthorized?
#500 opened Feb 10, 2025 by jspsiy
[Question] Training setup on remote Ubuntu server
#497 opened Feb 4, 2025 by lxbanov
Couldn't find any class folder in /val
#496 opened Jan 27, 2025 by jetic83
1
_parse_slurm_node_list leads to incorrect node names if no brackets are in node name
#495 opened Jan 27, 2025 by Leonngm
scaling down number of epoch to 1
#494 opened Jan 14, 2025 by clemsgrs
1
Dinov2 Traing error
#492 opened Jan 4, 2025 by wyh196646
1
INT8 Quantization of dinov2 TensorRT Model is Not Faster than FP16 Quantization
#489 opened Dec 6, 2024 by mr-lz
Unable to Replicate Depth Estimation Results on NYUv2 Depth Dataset (Including After Applying Eigen Crop)
#484 opened Nov 22, 2024 by erikjagnandan
1
linear layer input
#483 opened Nov 15, 2024 by kurbobo
Environment installation problem
#481 opened Nov 9, 2024 by wyh196646
7
Train ViT-g/14 on Imagenet-1k from scratch
#480 opened Nov 9, 2024 by QAQdev
if model is convergence. what range dose total loss in?Did someone find a set of hyper parameters adapt a small batch size?
#479 opened Nov 6, 2024 by risingClouds
5
Attention maps exhibit outliers even with registers
#475 opened Oct 25, 2024 by XiphosF
Regarding the loss issue of training from scratch on one's own dataset
#474 opened Oct 25, 2024 by DaIGaN2019
Checkpoint for ViT-L/14 Pretrained from Scratch (with SwiGLU)
#473 opened Oct 23, 2024 by dfan
Simple modification suggestion for the requirments.txt
#470 opened Oct 4, 2024 by jonghyunlee1993
1
DINOv2 models with registers lose the ability to semantically represent similar objects when performing PCA across images.
#469 opened Oct 2, 2024 by aenoca
2
Change align_corners to True in position encoding interpolation
#468 opened Sep 24, 2024 by infinity1096
2
Have you tried to include RoPE?
#467 opened Sep 21, 2024 by R-LEI2536
About the pre-trained model with patchsize=16
#466 opened Sep 19, 2024 by Hlll-huang
1
Access to pretrained weights of CLIP with register
#465 opened Sep 14, 2024 by niiickZ
Are registers and VPT-shallow identical?
#464 opened Sep 11, 2024 by qlOvQlp
ProTip! Follow long discussions with comments:>50.