yzslab / gaussian-splatting-lightning

A 3D Gaussian Splatting framework with various derived algorithms and an interactive web viewer
Other
457 stars 38 forks source link

viewer not working after training #3

Closed antoinebio closed 11 months ago

antoinebio commented 11 months ago

Hi

after a first train I want to view the trained model

but python viewer.py script used like that python .\viewer.py .\outputs\TEST01\ returns that issue below...

image

Is my train folder consitent ?

image

yzslab commented 11 months ago

As your image shown, the training was interrupted. Please try again after training finished.

antoinebio commented 11 months ago

@yzslab is it possible to limit the train to a limit value for good end of that step?

yzslab commented 11 months ago

@yzslab is it possible to limit the train to a limit value for good end of that step?

By default, the training will stop at 30k iterations. You can change this value by --max_steps options.

antoinebio commented 11 months ago

unfortunately it's not working

end of iteration here ( i set 100 iterations)

image

and viewer launch...

image

yzslab commented 11 months ago

unfortunately it's not working

end of iteration here ( i set 100 iterations)

image

and viewer launch...

image

Run this:

pip uninstall viser
pip install viser==0.1.9
antoinebio commented 11 months ago

thanks that solved the last issue.

how does viser work ? the viewer part ?

should I install and use httpserver to emulate the server , or XAMPP ?

image

I tried the URL in a chrome browser without any

should I use that nerfstrudio viewer instead ?

https://viewer.nerf.studio/?websocket_url=ws://localhost:7007

yzslab commented 11 months ago

thanks that solved the last issue.

how does viser work ? the viewer part ?

should I install and use httpserver to emulate the server , or XAMPP ?

image

I tried the URL in a chrome browser without any

should I use that nerfstrudio viewer instead ?

https://viewer.nerf.studio/?websocket_url=ws://localhost:7007

Access: http://localhost:8080