zip nerf is not working 360.gin file is missing

[sumanttyagi]

as soon we start to train a model after processing an images with comlap.

below isssues are coming

ERROR:root:Path not found: /configs/360.gin
Traceback (most recent call last):
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\runpy.py", line 194, in _run_module_as_main
    return _run_code(code, main_globals, None,
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\runpy.py", line 87, in _run_code
    exec(code, run_globals)
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\Scripts\ns-train.exe\__main__.py", line 7, in <module>
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\nerfstudio\scripts\train.py", line 262, in entrypoint
    main(
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\nerfstudio\scripts\train.py", line 247, in main
    launch(
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\nerfstudio\scripts\train.py", line 189, in launch    main_func(local_rank=0, world_size=world_size, config=config)
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\nerfstudio\scripts\train.py", line 99, in train_loop
    trainer.setup()
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\nerfstudio\engine\trainer.py", line 153, in setup    self.pipeline = self.config.pipeline.setup(
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\nerfstudio\configs\base_config.py", line 54, in setup
    return self._target(self, **kwargs)
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\zipnerf_ns\zipnerf_pipeline.py", line 63, in __init__
    self._model = config.model.setup(
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\nerfstudio\configs\base_config.py", line 54, in setup
    return self._target(self, **kwargs)
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\nerfstudio\models\base_model.py", line 85, in __init__
    self.populate_modules()  # populate the modules
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\zipnerf_ns\zipnerf_model.py", line 56, in populate_modules
    gin.parse_config_files_and_bindings(gin_files, None)
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\gin\config.py", line 2497, in parse_config_files_and_bindings
    includes_and_imports = parse_config_file(config_file, skip_unknown)
  File "C:\Users\sum11525\Anaconda3\envs\nerfstudio1\lib\site-packages\gin\config.py", line 2457, in parse_config_file
    raise IOError(err_str.format(config_file, prefixes))
OSError: Unable to open file: /configs/360.gin. Searched config paths: [''].

[sumanttyagi]

this issue was reported earlier no enhancement or fix is provided till now -https://github.com/nerfstudio-project/nerfstudio/issues/2923

[ichsan2895]

this issue was reported earlier no enhancement or fix is provided till now -https://github.com/nerfstudio-project/nerfstudio/issues/2923

Have you try this way? It works with Linux. I never test on Windows

pip install nerfstudio==1.0.3
pip install git+https://github.com/NVlabs/tiny-cuda-nn/#subdirectory=bindings/torch

git clone https://github.com/SuLvXiangXin/zipnerf-pytorch.git
cd zipnerf-pytorch
pip install ./extensions/cuda
pip install torch-scatter -f https://data.pyg.org/whl/torch-2.0.0+${CUDA}.html

pip install -e . 
ns-install-cli

Then you will able to run this command

ns-train zipnerf nerfstudio-data --data path/to/your/data

[sumanttyagi]

@ichsan2895 Please check wil running pip install ./extensions/cuda

Processing ./extensions/cuda
  Preparing metadata (setup.py) ... done
Building wheels for collected packages: cuda-backend
  error: subprocess-exited-with-error
  
  × python setup.py bdist_wheel did not run successfully.
  │ exit code: 1
  ╰─> See above for output.
  
  note: This error originates from a subprocess, and is likely not a problem with pip.
  Building wheel for cuda-backend (setup.py) ... error
  ERROR: Failed building wheel for cuda-backend
  Running setup.py clean for cuda-backend
Failed to build cuda-backend
ERROR: ERROR: Failed to build installable wheels for some pyproject.toml based projects (cuda-backend)

[koktavy]

@sumanttyagi I ran into that error during Zipnerf install, and was able to resolve that error by setting the C++ version:

set NVCC_APPEND_FLAGS=-std=c++17

But trying to install torch-scatter broke my installation, possibly because of torch 2.0 (in)compatibility.

I still have the /configs/360.gin error.

[koktavy]

Ah, I was able to provide the 360.gin file manually and start training!

It's looking for /configs/360.gin relative to the command directory. So if you're running your command from the root folder of nerfstudio like I am, create the configs folder in nerfstudio. We'll see how the training goes 🤞

Next step for me will be to run the Reality Capture poses through Zip NeRF instead of the COLMAP pose data, if possible.

[sumanttyagi]

@koktavy

• [ ] it should come from the repo the 360.gin file this is pure bug or this steps should be mentioned in repo

• [ ] and also can you explain each parameter of 360.gin file how to set it for custom dataset to train model

• [ ] [THIS WORKED FOR ME ]also the above bug is linked to something else for me Uninstall the ninja library then reinstall the process at last do pip install ninja this is the know issue mentioned on other or parent repo STILL THE 360.GIN FILE WILL BE MISSING BUT env will be created

[sumanttyagi]

• [ ] Also with manually placing the 360.gin file is giving very poor results clouds like noise everywhere in nerf model please check if it coming from original model

[koktavy]

I agree, it should be included, something is not right.

After training "finished" the results did not converge. I'm not familiar with the GIN file and what the parameters control, I was hoping it would be set to ideal defaults.

With the env created are you able to train successfully or you have the same error?

[sumanttyagi]

@koktavy not working please revisit the code