whether or not to install cudatoolkit-dev

[Ericgone]

Hi,

thanks for you greak work. but your installation instructions doesn't include installing cudatoolkit-dev. I wonder if it was mistakenly excluded by you. If yes, which version matches your project? best,

[zhengzhang01]

For this project, the version of cudatoolkit-dev that matches the specified PyTorch and CUDA versions is cudatoolkit-dev=11.8.

Here are the updated installation instructions:

conda create -n pixelgs python=3.9 -y

conda install pytorch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 pytorch-cuda=11.8 -c pytorch -c nvidia

conda install -c nvidia cudatoolkit-dev=11.8

pip install -r requirements.txt

cd submodules pip install ./diff-gaussian-rasterization pip install ./simple-knn

[Ericgone]

For this project, the version of cudatoolkit-dev that matches the specified PyTorch and CUDA versions is cudatoolkit-dev=11.8.

Here are the updated installation instructions:

conda create -n pixelgs python=3.9 -y

conda install pytorch==2.0.1 torchvision==0.15.2 torchaudio==2.0.2 pytorch-cuda=11.8 -c pytorch -c nvidia

conda install -c nvidia cudatoolkit-dev=11.8

pip install -r requirements.txt

cd submodules pip install ./diff-gaussian-rasterization pip install ./simple-knn

thanks very much for your prompt reply. but the command "conda install -c nvidia cudatoolkit-dev=11.8" still doesn't work since the package 11.8 was not found. you can try the command to see if it works by creating a anaconda virtual environment. I guess you will encounter with the same problem. best,

[zhengzhang01]

Below is the environment.yml file:

name: pixelgs
channels:
  - pytorch
  - nvidia
  - conda-forge
  - defaults
dependencies:
  - python=3.9
  - cudatoolkit=11.8
  - pip
  - pytorch=2.0.1
  - torchaudio=2.0.2
  - torchvision=0.15.2
  - tqdm
  - plyfile=0.8.1
  - GPUtil
  - pip:
    - submodules/diff-gaussian-rasterization
    - submodules/simple-knn

You can try using the following commands to set up the environment:

conda env create --file environment.yml
conda activate pixelgs