Closed xiyufeng2 closed 5 months ago
Hi, this appears to be an issue with your gsplat
installation, or an issue with Nerfstudio's Gaussian Splat.
Does ns-train splatfacto ...
work for you? Otherwise, I would try reinstalling the gsplat
package, perhaps from the GitHub repo.
Hello
Adding to this thread, similar issue but different error
When I try to run GARField with Gaussian Splatting, I get the following error:
RuntimeError: Optimizer config for 'xyz' not found in config file. Make sure you specify an optimizer for each parameter group. Provided configs were: dict_keys(['means', 'features_dc', 'features_rest', 'opacities', 'scales', 'quats', 'camera_opt'])
To answer the question above, yes ns-train splatfacto
works for me
Just to check I've got the arguments right, my command is:
ns-train garfield-gauss --data <path to original data used for Garfield training> --pipeline.garfield-ckpt <path to config.yml from Garfield training>
thanks in advance
Hi, the parameter name mismatch issue is due to the recent splatfacto rename PR in nerfstudio (https://github.com/nerfstudio-project/nerfstudio/pull/2946), which this repo addresses in https://github.com/chungmin99/garfield/pull/23. This issue is related to https://github.com/chungmin99/garfield/issues/26.
garfield-gauss
should work if you pull the latest version of nerfstudio.
Hi
Many thanks for the reply I did look through the other issues, but missed that one
Andrew
The previous process in secceed for ''Running GARField''. But sth. went wrong when I use the command ''ns-train garfield-gauss --data /your/data/here --pipeline.garfield-ckpt outputs/y/.../config.yml''. Error is as following:
[5/5] c++ bindings.cuda.o backward.cuda.o forward.cuda.o ext.o -shared -L/home/moonlight/.conda/envs/nerfstudio/lib/python3.8/site-packages/torch/lib -lc10 -lc10_cuda -ltorch_cpu -ltorch_cuda -ltorch -ltorch_python -L/home/moonlight/.conda/envs/nerfstudio/lib64 -lcudart -o gsplat_cuda.so FAILED: gsplat_cuda.so c++ bindings.cuda.o backward.cuda.o forward.cuda.o ext.o -shared -L/home/moonlight/.conda/envs/nerfstudio/lib/python3.8/site-packages/torch/lib -lc10 -lc10_cuda -ltorch_cpu -ltorch_cuda -ltorch -ltorch_python -L/home/moonlight/.conda/envs/nerfstudio/lib64 -lcudart -o gsplat_cuda.so /usr/bin/ld: cannot find -lcudart: No such file or directory collect2: error: ld returned 1 exit status ninja: build stopped: subcommand failed.
can you afford some help?thank you