Warnings while running on an SSH session - no output videos about 3d-photo-inpainting

[MahadevanSrinivasan]

Thanks for your reply. I did that and get the following error now:

Traceback (most recent call last):
  File "main.py", line 28, in <module>
    vispy.use(app='egl')
  File "/home/md/anaconda2/envs/3DP/lib/python3.7/site-packages/vispy/util/wrappers.py", line 97, in use
    use_app(app)
  File "/home/md/anaconda2/envs/3DP/lib/python3.7/site-packages/vispy/app/_default_app.py", line 47, in use_app
    default_app = Application(backend_name)
  File "/home/md/anaconda2/envs/3DP/lib/python3.7/site-packages/vispy/app/application.py", line 49, in __init__
    self._use(backend_name)
  File "/home/md/anaconda2/envs/3DP/lib/python3.7/site-packages/vispy/app/application.py", line 235, in _use
    raise RuntimeError(msg)
RuntimeError: Could not import backend "EGL":
Could not initialize

[MahadevanSrinivasan]

@LemonATsu I checked with Mac and linux different devices but stuck here.

Traceback (most recent call last):
File "main.py", line 68, in
map_location=torch.device(device))
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 529, in load
return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 702, in _legacy_load
result = unpickler.load()
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 665, in persistent_load
deserialized_objects[root_key] = restore_location(obj, location)
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 740, in restore_location
return default_restore_location(storage, str(map_location))
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 156, in default_restore_location
result = fn(storage, location)
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 132, in _cuda_deserialize
device = validate_cuda_device(location)
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 116, in validate_cuda_device
raise RuntimeError('Attempting to deserialize object on a CUDA '
RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.

Set the gpus field in the arugment.yml file to -1 to fallback to CPU

[LemonATsu]

Try setting offscreen_rendering in argument.yml to True to enable rendering without GUI environment.

[LemonATsu]

Try installing EGL by running: sudo apt-get install libgles2-mesa-dev. Let us know if it works. Thanks

[LemonATsu]

@cahitatlas Try run the download.sh script to download all the pre-trained weights.

[puppet101]

@LemonATsu
I get the same error, but I am using CentOS 7, what packages should I install to run it with offscreen rendering?

[ShihMengLi]

Maybe this one?
https://centos.pkgs.org/7/centos-x86_64/mesa-libGLU-devel-9.0.0-4.el7.x86_64.rpm.html

[MahadevanSrinivasan]

Try installing EGL by running: sudo apt-get install libgles2-mesa-dev. Let us know if it works. Thanks

I have that package installed already. But still see the same error

[sdan]

I've been tackling this issue as well.
I don't see the reason for having all those render() functions in mesh.py. At this point it seems like xvfb-run is the best temporary option.

The weird thing about this is:
It runs perfectly on Colab but on a Docker container or even another Jupyter notebook these vispy issues come up

[sdan]

The issue here is: regardless of offscreen_render, you will still need vispy's scene module to create canvases for mesh.py to run.

I've installed all the packages but strangely I can't replicate this anywhere other than colab (when I run this on a jupyter notebook I run into "no screen" errors

[MoePus]

Try setting offscreen_rendering in argument.yml to True to enable rendering without GUI environment.

This works for win10 wsl unbuntu 18.04

[dilenpant]

@LemonATsu I checked with Mac and linux different devices but stuck here.

Traceback (most recent call last):
File "main.py", line 68, in
map_location=torch.device(device))
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 529, in load
return _legacy_load(opened_file, map_location, pickle_module, **pickle_load_args)
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 702, in _legacy_load
result = unpickler.load()
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 665, in persistent_load
deserialized_objects[root_key] = restore_location(obj, location)
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 740, in restore_location
return default_restore_location(storage, str(map_location))
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 156, in default_restore_location
result = fn(storage, location)
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 132, in _cuda_deserialize
device = validate_cuda_device(location)
File "/home/dilendra/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 116, in validate_cuda_device
raise RuntimeError('Attempting to deserialize object on a CUDA '
RuntimeError: Attempting to deserialize object on a CUDA device but torch.cuda.is_available() is False. If you are running on a CPU-only machine, please use torch.load with map_location=torch.device('cpu') to map your storages to the CPU.

[MahadevanSrinivasan]

Try setting offscreen_rendering in argument.yml to True to enable rendering without GUI environment.

This works for win10 wsl unbuntu 18.04

Did you have to install any new packages? I tried on Win10 WSL Ubuntu 16.04 and it does not work.

[MoePus]

Try setting offscreen_rendering in argument.yml to True to enable rendering without GUI environment.

This works for win10 wsl unbuntu 18.04

Did you have to install any new packages? I tried on Win10 WSL Ubuntu 16.04 and it does not work.
Only libgles2-mesa-dev.
My wsl ubuntu is new installed on win10 1909.
And i have aslo run export DISPLAY=:0

[davideandres95]

Try setting offscreen_rendering in argument.yml to True to enable rendering without GUI environment.

This works for win10 wsl unbuntu 18.04

Well it does not work in ubuntu 16.04. Any other suggestions?

[cahitatlas]

I get same problem on macOS Catalina
Normaly when i try with default yml file, i get that result:

(3DP) bash-3.2$ python main.py --config argument.yml
0%| | 0/1 [00:00<?, ?it/s]Current Source ==> einstein
initialize
device: cpu
0%| | 0/1 [00:00<?, ?it/s]
Traceback (most recent call last):
File "main.py", line 44, in
config['MiDaS_model_ckpt'], MonoDepthNet, MiDaS_utils, target_w=640)
File "/Users/cahitatlas/Downloads/3d-photo-inpainting-master/MiDaS/run.py", line 27, in run_depth
model = Net(model_path)
File "/Users/cahitatlas/Downloads/3d-photo-inpainting-master/MiDaS/monodepth_net.py", line 52, in init
self.load(path)
File "/Users/cahitatlas/Downloads/3d-photo-inpainting-master/MiDaS/monodepth_net.py", line 88, in load
parameters = torch.load(path)
File "/Users/cahitatlas/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 584, in load
with _open_file_like(f, 'rb') as opened_file:
File "/Users/cahitatlas/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 234, in _open_file_like
return _open_file(name_or_buffer, mode)
File "/Users/cahitatlas/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 215, in init
super(_open_file, self).init(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: 'MiDaS/model.pt'

I deleted MiDaS_model_ckpt:
than i get result :

Start Running 3D_Photo ...
0%| | 0/1 [00:13<?, ?it/s]
Traceback (most recent call last):
File "main.py", line 66, in
map_location=torch.device(device))
File "/Users/cahitatlas/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 584, in load
with _open_file_like(f, 'rb') as opened_file:
File "/Users/cahitatlas/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 234, in _open_file_like
return _open_file(name_or_buffer, mode)
File "/Users/cahitatlas/anaconda3/envs/3DP/lib/python3.7/site-packages/torch/serialization.py", line 215, in init
super(_open_file, self).init(open(name, mode))
FileNotFoundError: [Errno 2] No such file or directory: 'checkpoints/edge-model.pth'

[lizhengwei1992]

Try installing EGL by running: sudo apt-get install libgles2-mesa-dev. Let us know if it works. Thanks

I also have the same error "RuntimeError: Could not import backend "EGL":
" after install libgles2-mesa-dev.

[lizhengwei1992]

Try installing EGL by running: sudo apt-get install libgles2-mesa-dev. Let us know if it works. Thanks

I have that package installed already. But still see the same error

Use os.environ['DISPLAY']= ':0' in main.py after running: sudo apt-get install libgles2-mesa-dev, fixed it !

[LemonATsu]

@lizhengwei1992 Thanks for letting us know. This would be useful for others that have the same problem.

[mohit1997]

This worked for me while rendering videos on a server,

I specifically changed two things,

On the client-side while logging in, use

ssh -v -X user@server

then in the argument.yml file update

offscreen_rendering: True

[yfq512]

I solved same error follow this:
vispy/vispy#1848 (comment)