search

SHI-Labs / OneFormer

OneFormer: One Transformer to Rule Universal Image Segmentation, arxiv 2022 / CVPR 2023
https://praeclarumjj3.github.io/oneformer
MIT License
1.44k stars 130 forks source link

Is it possible to run the code without wandb? #7

Closed asmith26 closed 1 year ago

asmith26 commented 1 year ago

Hi there, thanks very much for publishing this repo, it looks very interesting.

I'm trying to follow the installation instructions but due to the CPU architecture of the system I'm using, I don't think I'll be able to use wandb (I don't have an account with them so thought I'd try run it locally):

$ wandb server start
WARNING: The requested image's platform (linux/amd64) does not match the detected host platform (linux/ppc64le) and no specific platform was requested

Is it possible to run the code without wandb?

Thanks for any help! :)

praeclarumjj3 commented 1 year ago

Hi @asmith26, thanks for your interest in our work.

If you want to run the code without wandb, you need to remove the following lines in train_net.py.

https://github.com/SHI-Labs/OneFormer/blob/d20a596fef5568e2c19bd14eaa208fb2896e772d/train_net.py#L191 https://github.com/SHI-Labs/OneFormer/blob/d20a596fef5568e2c19bd14eaa208fb2896e772d/train_net.py#L397-L398

Robotatron commented 1 year ago

Removing those WandB dependencies works, thanks @praeclarumjj3

But there is no log data for tensorboard. I am wondering was tensorboard logging disabled for Oneformer on purpose? Usually Detectron2, and models build on top of it (Mask2Former, MaskDINO, etc) all log tensorboard data by default.

Is there an easy way to bring tensorboard logging back or will it require heavy source code modifications of OneFormer?

praeclarumjj3 commented 1 year ago

Hi @Robotatron, to avoid redundancy, we do not use tensorboard logging as we already use wandb logging. You can re-enable Tensorboard logging with minimal changes: