Repo link for diff-gaussian-rasterization-w-depth was not working using yml.
The file doesn't have cuda-toolkit dependency but it's written in the instruction.
I also met an issue when compiling gaussian splatting due to incompatable gcc&g++ version with CUDA. The reason here is because I was using gcc14 and according to https://gist.github.com/ax3l/9489132 CUDA 11.6 doesn't support gcc and g++ greater than certain versions. I fixed it by downgrading my gcc and g++ to version 10.x
conda install gxx_linux-64=10
LOG
stdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -DTORCH_EXTENSION_NAME=_C -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 -std=c++14
cuda_rasterizer/auxiliary.h(151): warning #177-D: variable "p_proj" was declared but never referenced
/home/adamliu/anaconda3/envs/splatam/bin/nvcc -I/home/adamliu/anaconda3/envs/splatam/lib/python3.10/site-packages/torch/include -I/home/adamliu/anaconda3/envs/splatam/lib/python3.10/site-packages/torch/include/torch/csrc/api/include -I/home/adamliu/anaconda3/envs/splatam/lib/python3.10/site-packages/torch/include/TH -I/home/adamliu/anaconda3/envs/splatam/lib/python3.10/site-packages/torch/include/THC -I/home/adamliu/anaconda3/envs/splatam/include -I/home/adamliu/anaconda3/envs/splatam/include/python3.10 -c cuda_rasterizer/rasterizer_impl.cu -o build/temp.linux-x86_64-cpython-310/cuda_rasterizer/rasterizer_impl.o -D__CUDA_NO_HALF_OPERATORS__ -D__CUDA_NO_HALF_CONVERSIONS__ -D__CUDA_NO_BFLOAT16_CONVERSIONS__ -D__CUDA_NO_HALF2_OPERATORS__ --expt-relaxed-constexpr --compiler-options '-fPIC' -I/tmp/pip-req-build-14he5j7u/third_party/glm/ -DTORCH_API_INCLUDE_EXTENSION_H -DPYBIND11_COMPILER_TYPE=\"_gcc\" -DPYBIND11_STDLIB=\"_libstdcpp\" -DPYBIND11_BUILD_ABI=\"_cxxabi1011\" -DTORCH_EXTENSION_NAME=_C -D_GLIBCXX_USE_CXX11_ABI=0 -gencode=arch=compute_86,code=compute_86 -gencode=arch=compute_86,code=sm_86 -std=c++14
cuda_rasterizer/auxiliary.h(151): warning #177-D: variable "p_proj" was declared but never referenced
/usr/include/c++/11/bits/std_function.h:435:145: error: parameter packs not expanded with ‘...’:
435 | function(_Functor&& __f)
| ^
/usr/include/c++/11/bits/std_function.h:435:145: note: ‘_ArgTypes’
/usr/include/c++/11/bits/std_function.h:530:146: error: parameter packs not expanded with ‘...’:
530 | operator=(_Functor&& __f)
| ^
/usr/include/c++/11/bits/std_function.h:530:146: note: ‘_ArgTypes’
error: command '/home/adamliu/anaconda3/envs/splatam/bin/nvcc' failed with exit code 1
[end of output]
note: This error originates from a subprocess, and is likely not a problem with pip.
ERROR: Failed building wheel for diff-gaussian-rasterization
Running setup.py clean for diff-gaussian-rasterization
Failed to build diff-gaussian-rasterization
ERROR: Could not build wheels for diff-gaussian-rasterization, which is required to install pyproject.toml-based projects
I also met an issue when compiling gaussian splatting due to incompatable gcc&g++ version with CUDA. The reason here is because I was using gcc14 and according to https://gist.github.com/ax3l/9489132 CUDA 11.6 doesn't support gcc and g++ greater than certain versions. I fixed it by downgrading my gcc and g++ to version 10.x
LOG