-
Notifications
You must be signed in to change notification settings - Fork 8
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.
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
Hello author, may I ask why there is no stage1.sh file in the tools file #7
Comments
Thanks for your attantion. Stage1 is trained on two machines, each equipped with eight V100 or A100 GPUs, and the code supports PyTorch's native multi-node and multi-GPU parallelism. If you want to train Stage1, you simply need to modify the configuration file for Stage2 located in the tools directory. Here are the steps: Step 1: Configure the parameters for multi-node and multi-GPU parallelism. For detailed instructions, refer to Detectron2's launch script documentation. Step 2: Change the configuration file from --config-file configs/univs/univs_r50_stage2.yaml to --config-file configs/univs/univs_r50_stage1.yaml. |
Dear Zhao,
You can find more detailed instructionz from
https://github.com/MinghanLi/UniVS/blob/main/datasets/README.md . Thanks.
Best,
Minghan
zhao ***@***.***> 于2024年5月16日周四 17:07写道:
… Again, I would like to inform
cfg. MODEL. UniVS.CLIP_CLASS_EMBED_PATH =
'datasets/concept_emb/combined_datasets_cls_emb_rn50x4.pth' where should I
download the weight file here, thank you very much!
—
Reply to this email directly, view it on GitHub
<#7 (comment)>, or
unsubscribe
<https://github.com/notifications/unsubscribe-auth/AHOANLQGMTTA4WBXTTXT5RLZCRZNZAVCNFSM6AAAAABHVI2UUWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMZDCMJUGYYTQOBUG4>
.
You are receiving this because you commented.Message ID:
***@***.***>
|
Thanks to the author for the reply and I wish you all the best in your work! |
No description provided.
The text was updated successfully, but these errors were encountered: