NSVF Neural Sparse Voxel Fields – NeRF Neural Radiance Fields

NSVF ! (Updated) Not working on windows 24102020 wsl1

Windows 10 Anaconda

conda create -n nsvf python=3.6
conda activate nsvf
conda install pytorch torchvision cudatoolkit=10.2 -c pytorch
pip install -r requirements.txt

 Hi, we have not tested on Windows 10 yet. Our code is based on cuda, so it is not possible to use wsl1. Recently wsl2 supports cuda now, but I have never successfully tested it before.
 Our code now does not rely on torchsearchsorted now. Maybe you can delete the requirement for that and try again for installing on windows? 

2020-10-26 17:11:19 | INFO | fairnr_cli.train | model nsvf_base, criterion SRNLossCriterion
 2020-10-26 17:11:19 | INFO | fairnr_cli.train | num. model params: 582737 (num. trained: 582724)
 2020-10-26 17:11:22 | INFO | fairseq.utils | CUDA enviroments for all 1 workers
 2020-10-26 17:11:22 | INFO | fairseq.utils | rank   0: capabilities =  7.5  ; total memory = 6.000 GB ; name = GeForce RTX 2060
 2020-10-26 17:11:22 | INFO | fairseq.utils | CUDA enviroments for all 1 workers
 2020-10-26 17:11:22 | INFO | fairnr_cli.train | training on 1 GPUs
 2020-10-26 17:11:22 | INFO | fairnr_cli.train | max tokens per GPU = None and max sentences per GPU = 1
 2020-10-26 17:11:22 | INFO | fairseq.trainer | no existing checkpoint found checkpoint\Wineholder\nsvf_basev1\checkpoint_last.pt
 2020-10-26 17:11:22 | INFO | fairseq.trainer | loading train data for epoch 1

Windows 10 WSL

speedoo@SPEEDOO:~/NSVF$ python3 -m pip install --upgrade pip
speedoo@SPEEDOO:~/NSVF$ sudo pip install ninja

Nvidia Toolkit
wget https://developer.download.nvidia.com/compute/cuda/repos/wsl-ubuntu/x86_64/cuda-wsl-ubuntu.pin

sudo mv cuda-wsl-ubuntu.pin /etc/apt/preferences.d/cuda-repository-pin-600

sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/wsl-ubuntu/x86_64/7fa2af80.pub

sudo add-apt-repository "deb https://developer.download.nvidia.com/compute/cuda/repos/wsl-ubuntu/x86_64/ /"

sudo apt-get updatesudo apt-get -y install cuda

PyTorch
pip install torch==1.6.0+cu101 torchvision==0.7.0+cu101 -f https://download.pytorch.org/whl/torch_stable.html

speedoo@SPEEDOO:~/NSVF$ git clone https://github.com/facebookresearch/NSVF.git
speedoo@SPEEDOO:~/NSVF$ cd NSVF

OK sudo pip install torch torchvision --use-feature=2020-resolver
OU
sudo pip install torch===1.6.0 torchvision===0.7.0 -f https://download.pytorch.org/whl/torch_stable.html
sudo pip install torch==1.5.1+cu101 torchvision==0.6.1+cu101 -f https://download.pytorch.org/whl/torch_stable.html

speedoo@SPEEDOO:~/NSVF$ pip install -r requirements.txt --use-feature=2020-resolver

speedoo@SPEEDOO:~/NSVF$ python3 -c "import torch; print(torch.version.cuda)"
 10.2
speedoo@SPEEDOO:~/NSVF$ sudo apt install nvidia-cuda-toolkit

speedoo@SPEEDOO:~/NSVF$ sudo nvcc --version
 nvcc: NVIDIA (R) Cuda compiler driver
 Copyright (c) 2005-2019 NVIDIA Corporation
 Built on Sun_Jul_28_19:07:16_PDT_2019
 Cuda compilation tools, release 10.1, V10.1.243

1)  python3 setup.py build_ext --inplace
ou 2) pip install --editable ./

Redo speedoo@SPEEDOO:~/NSVF$ pip install -r requirements.txt

Waiting for An improved version of NeRF: Neural Sparse Voxel Fields

Using :

https://github.com/kwea123/nerf_pl

https://github.com/kwea123/nerf_Unity

https://colmap.github.io/index.html