首先，创建一个环境，Python版本为3.10

conda create -n pixel_splat python=3.10 -y

接着，激活这个环境

conda activate pixel_splat

下一步安装torch, torchaudio, torchvision.

pip install torch==2.7.1 --index-url https://download.pytorch.org/whl/cu118

pip install torchaudio==2.7.1 --index-url https://download.pytorch.org/whl/cu118

pip install torchvision==0.22.1 --index-url https://download.pytorch.org/whl/cu118

镜像安装：

pip install torch==2.7.1 torchaudio==2.7.1 torchvision==0.22.1 -f https://mirrors.tuna.tsinghua.edu.cn/pytorch-wheels/cu118 -i https://pypi.tuna.tsinghua.edu.cn/simple

如果得到Successfully installed字样，忽略其他包is not installed的这类错误

如果是用镜像安装的torch，直接跳到安装diffusion-gaussian-rasterization这一步。

退出这个虚拟环境

conda deactivate

再下一步，用这个yml文件更新这个环境，记得在yml文件中修改虚拟环境的名称，和之前创建的环境一致

conda env update -f environment.yml

yml文件内容如下：

name: pixel_splat
channels:
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/cloud/pytorch/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/free/
  - https://mirrors.tuna.tsinghua.edu.cn/anaconda/pkgs/main/
  - bioconda
  - defaults
dependencies:
  - _libgcc_mutex=0.1=main
  - _openmp_mutex=5.1=1_gnu
  - bzip2=1.0.8=h5eee18b_6
  - ca-certificates=2026.3.19=h06a4308_0
  - ld_impl_linux-64=2.44=h9e0c5a2_3
  - libexpat=2.7.5=h7354ed3_0
  - libffi=3.4.4=h6a678d5_1
  - libgcc=15.2.0=h69a1729_7
  - libgcc-ng=15.2.0=h166f726_7
  - libgomp=15.2.0=h4751f2c_7
  - libnsl=2.0.0=h5eee18b_0
  - libstdcxx=15.2.0=h39759b7_7
  - libstdcxx-ng=15.2.0=hc03a8fd_7
  - libuuid=1.41.5=h5eee18b_0
  - libxcb=1.17.0=h9b100fa_0
  - libzlib=1.3.1=hb25bd0a_0
  - ncurses=6.5=h7934f7d_0
  - openssl=3.5.5=h1b28b03_0
  - packaging=26.0=py310h06a4308_0
  - pip=26.0.1=pyhc872135_1
  - pthread-stubs=0.3=h0ce48e5_1
  - python=3.10.20=h741d88c_0
  - readline=8.3=hc2a1206_0
  - sqlite=3.51.2=h3e8d24a_0
  - tk=8.6.15=h54e0aa7_0
  - tzdata=2026a=he532380_0
  - wheel=0.46.3=py310h06a4308_0
  - xorg-libx11=1.8.12=h9b100fa_1
  - xorg-libxau=1.0.12=h9b100fa_0
  - xorg-libxdmcp=1.1.5=h9b100fa_0
  - xorg-xorgproto=2024.1=h5eee18b_1
  - xz=5.8.2=h448239c_0
  - zlib=1.3.1=hb25bd0a_0
  - pip:
      - aiohappyeyeballs==2.6.1
      - aiohttp==3.13.5
      - aiosignal==1.4.0
      - annotated-doc==0.0.4
      - annotated-types==0.7.0
      - antlr4-python3-runtime==4.9.3
      - anyio==4.13.0
      - appdirs==1.4.4
      - async-timeout==5.0.1
      - attrs==26.1.0
      - beartype==0.14.1
      - black==26.3.1
      - certifi==2026.2.25
      - charset-normalizer==3.4.7
      - click==8.3.2
      - colorama==0.4.6
      - colorspacious==1.1.2
      - conda-pack==0.9.1
      - contourpy==1.3.2
      - cycler==0.12.1
      - dacite==1.9.2
      - decorator==4.4.2
      - docker-pycreds==0.4.0
      - e3nn==0.6.0
      - einops==0.8.2
      - exceptiongroup==1.3.1
      - filelock==3.25.2
      - fonttools==4.62.1
      - frozenlist==1.8.0
      - fsspec==2026.2.0
      - gitdb==4.0.12
      - gitpython==3.1.46
      - h11==0.16.0
      - hf-xet==1.4.3
      - httpcore==1.0.9
      - httpx==0.28.1
      - huggingface-hub==1.9.0
      - hydra-core==1.3.2
      - idna==3.11
      - imageio==2.37.3
      - imageio-ffmpeg==0.6.0
      - jaxtyping==0.2.19
      - jinja2==3.1.6
      - kiwisolver==1.5.0
      - lazy-loader==0.5
      - lightning==2.6.1
      - lightning-utilities==0.15.3
      - lpips==0.1.4
      - markdown-it-py==4.0.0
      - markupsafe==3.0.3
      - matplotlib==3.10.8
      - mdurl==0.1.2
      - moviepy==1.0.3
      - mpmath==1.3.0
      - multidict==6.7.1
      - mypy-extensions==1.1.0
      - networkx==3.4.2
      - numpy==1.26.4
      - nvidia-cublas-cu11==11.11.3.6
      - nvidia-cuda-cupti-cu11==11.8.87
      - nvidia-cuda-nvrtc-cu11==11.8.89
      - nvidia-cuda-runtime-cu11==11.8.89
      - nvidia-cudnn-cu11==9.1.0.70
      - nvidia-cufft-cu11==10.9.0.58
      - nvidia-curand-cu11==10.3.0.86
      - nvidia-cusolver-cu11==11.4.1.48
      - nvidia-cusparse-cu11==11.7.5.86
      - nvidia-nccl-cu11==2.21.5
      - nvidia-nvtx-cu11==11.8.86
      - omegaconf==2.3.0
      - opencv-python==4.6.0.66
      - opt-einsum==3.4.0
      - opt-einsum-fx==0.1.4
      - pathspec==1.0.4
      - pillow==11.3.0
      - platformdirs==4.9.4
      - plyfile==1.1.3
      - proglog==0.1.12
      - propcache==0.4.1
      - protobuf==4.25.9
      - psutil==7.2.2
      - pydantic==2.12.5
      - pydantic-core==2.41.5
      - pygments==2.20.0
      - pyparsing==3.3.2
      - python-dotenv==1.2.2
      - pytokens==0.4.1
      - pytorch-lightning==2.6.1
      - pytz==2026.2
      - pyyaml==6.0.3
      - requests==2.33.1
      - rich==14.3.3
      - ruff==0.15.9
      - safetensors==0.7.0
      - scikit-image==0.25.2
      - scipy==1.15.3
      - sentry-sdk==2.57.0
      - setproctitle==1.3.7
      - setuptools==69.5.1
      - shellingham==1.5.4
      - six==1.17.0
      - sk-video==1.1.10
      - smmap==5.0.3
      - svg-py==1.10.0
      - sympy==1.14.0
      - tabulate==0.10.0
      - tensorboard==2.20.0
      - tifffile==2025.5.10
      - timm==1.0.26
      - tomli==2.4.1
      - torchmetrics==1.9.0
      - tqdm==4.67.3
      - triton==3.3.1
      - typer==0.24.1
      - typing-extensions==4.15.0
      - typing-inspection==0.4.2
      - urllib3==2.6.3
      - wadler-lindig==0.1.7
      - wandb==0.16.2
      - yarl==1.23.0

接着，激活这个虚拟环境

conda activate pixel_splat

如果用镜像安装torch，直接跳到这一步。
最后，安装diffusion-gaussian-rasterization

pip install git+https://gitclone.com/github.com/dcharatan/diff-gaussian-rasterization-modified --no-build-isolation

注意：是https://gitclone.com/github.com/ 而非 https://github.com/

在环境中输入python，然后

import torch

import diff_gaussian_rasterization

看看有没有报错。
如果不是镜像安装的torch，就结束了。

如果是通过镜像安装torch的话，需要用mvsplat的requirement.txt文件更新环境

pip install -r requirements.txt -i https://pypi.tuna.tsinghua.edu.cn/simple --exists-action i