Skip to content
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

several training warnings and headless eval #10

Open
Holmes-GU opened this issue Nov 19, 2024 · 2 comments
Open

several training warnings and headless eval #10

Holmes-GU opened this issue Nov 19, 2024 · 2 comments

Comments

@Holmes-GU
Copy link

Dear authors,

Thanks for releasing codes, which helps me learn the field. I encounter several issues and do not know how to fixed them. It is appreciated if you could give some help.

  1. For the training of RLBench:
    Following your training instruction "python3 train.py config=./configs/arp_plus.yaml hydra.job.name=arp_plus train.num_gpus=2 train.bs=96", Two warning appear as shown in the figure attached. Do you have any idea to deal with them>
image
  1. For the RLBench evaluation:
    Since I use miniconda to manage different environments, it do not have the file 'sysroot' in "$HOME/anaconda/x86_64-conda-linux-gnu/". What should I do to run RLBench on a Headless Server without Sudo as shown in the image? If I have sudo permission, what should I do to run RLBench on a remote server?
image
@mlzxy
Copy link
Owner

mlzxy commented Nov 19, 2024

For the warnings, I just confirmed that I also encountered them. They seem to come from the MVT backbone and shall not affect training.

The conda install anaconda:: xorg-x11-server-xvfb-cos6-×86_64 command creates the sysroot folder in the server I use, if it does not in yours, you can try:

option-1: try $H/anaconda3/x86_64-conda_cos6-linux-gnu or similar paths

option-2: run find . | grep Xvfb in folders like x86_64-conda-linux-gnu, as conda may store the executable in other paths

If you have sudo, then things should become easy. You can run sudo apt install xorg to install an x server, set DISPLAY=:0 environment variables, and then run the evaluation in headless mode.

@Holmes-GU
Copy link
Author

For the warnings, I just confirmed that I also encountered them. They seem to come from the MVT backbone and shall not affect training.

The conda install anaconda:: xorg-x11-server-xvfb-cos6-×86_64 command creates the sysroot folder in the server I use, if it does not in yours, you can try:

option-1: try $H/anaconda3/x86_64-conda_cos6-linux-gnu or similar paths

option-2: run find . | grep Xvfb in folders like x86_64-conda-linux-gnu, as conda may store the executable in other paths

If you have sudo, then things should become easy. You can run sudo apt install xorg to install an x server, set DISPLAY=:0 environment variables, and then run the evaluation in headless mode.

Got it. Thanks for your detailed reply.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

2 participants