Skip to content
This repository has been archived by the owner on Oct 31, 2023. It is now read-only.

Issues: facebookresearch/WSL-Images

17 Open 4 Closed
17 Open 4 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

Model performance
#23 opened Feb 16, 2022 by Tortoise17
cuda out of memory
#22 opened Nov 5, 2021 by naeem7789
Error
#21 opened Nov 2, 2021 by naeem7789
404 error when pulling model from hub
#20 opened Oct 22, 2021 by saarimrahman
7
Significant numerical differences with torch.amp.autocast() compared with stock pretrained resnext101
#17 opened Nov 5, 2020 by farleylai
How to find the precision of the various layers when laoding the models resnext101 models
#16 opened Oct 7, 2020 by aurotripathy
how can I do the final hashtag prediction using the pretrained model?
#14 opened Jul 30, 2020 by vyaslkv
Pretrained hashtag prediction model
#13 opened Mar 30, 2020 by sylvainma
4
Is there anyone who implemented the ResNeXt WSL reproduction results in tensorflow2.x or tensorflow1.x，Thx!
#12 opened Nov 15, 2019 by QQ2737499951
About training set
#11 opened Oct 27, 2019 by JrPeng
Passing 'pretrained' parameter to torch.hub.load raises error
#10 opened Oct 21, 2019 by sipan17
How to solve ZeroDivisionError: float division by zero?
#9 opened Oct 10, 2019 by mobassir94
1
WSL models not fine-tuned for ImageNet
#8 opened Sep 5, 2019 by simenkf
7
OSError: [Errno 22] Invalid argument
#7 opened Aug 19, 2019 by rafis
4
x101-32*8d doesn't work well when I put it in cascade rcnn in mmdetection
#5 opened Jul 2, 2019 by yudongliu97
5
About performance in Cub-200 finetune.
#4 opened Jun 27, 2019 by narrowsnap
2
Will ResNeXt-101 32x4d be available?
#2 opened Jun 26, 2019 by fengfu-chris
1
ProTip! Type g i on any issue or pull request to go back to the issue listing page.