Skip to content

ENVIRONMENT.md

Latest commit

 

History

History
77 lines (64 loc) · 1.97 KB

ENVIRONMENT.md

File metadata and controls

77 lines (64 loc) · 1.97 KB

Environment Setup

We describe steps (in Linux command line) to setup the environment for SparseFusion.

Conda Environment

We install and setup a conda environment.

(optional) Install Conda

Required if conda not installed.

cd ~
wget https://repo.anaconda.com/miniconda/Miniconda3-latest-Linux-x86_64.sh
chmod +x Miniconda3-latest-Linux-x86_64.sh
./Miniconda3-latest-Linux-x86_64.sh
export PATH="/home/username/miniconda/bin:$PATH"
conda init
source ~/.bashrc

Create New Environment

conda create -n sparsefusion python=3.8
conda activate sparsefusion

Install Dependencies

We install the necessary dependencies.

GCC and Cuda

Make sure to do this first!

We also assume that nvidia drivers and cuda=11.3.x is installed.

conda install -c conda-forge cxx-compiler=1.3.0
conda install -c conda-forge cudatoolkit-dev
conda install -c conda-forge ninja

Python Libraries

conda install pytorch==1.11.0 torchvision==0.12.0 torchaudio==0.11.0 cudatoolkit=11.3 -c pytorch
conda install -c fvcore -c iopath -c conda-forge fvcore iopath
conda install -c pytorch3d pytorch3d

Support Stable Diffusion

pip install transformers==4.19.2 pytorch-lightning==1.4.2 torchmetrics==0.6.0
pip install -e git+https://github.com/CompVis/taming-transformers.git@master#egg=taming-transformers

(optional) Install CO3D

Required if using full CO3D dataset.

git clone https://github.com/facebookresearch/co3d
cd co3d
pip install -r requirements.txt
pip install -e .

Install Other SparseFusion Requirements

cd sparsefusion
pip install -r requirements.txt

Building Extensions

We require a few extensions from torch-ngp. We detail how to install them below. See more details on the torch-ngp Github.

Build gridencoder

pip install ./external/gridencoder

Build raymarcher

pip install ./raymarching