Installation

😨 This Wiki is depreciated 😨 Visit the new Wiki at sapsan-wiki.github.io

Depreciated Installation

Sapsan can be run on both cpu and gpu. Below are the requirements for each version


CPU	torch>=1.9.0	torchvision>=0.10.0
GPU	torch>=1.9.0+cu111	torchvision>=0.10.0+cu111

Please follow the instructions on PyTorch to install either version. CUDA>=11.1 can be installed directly with PyTorch as well.

pip install sapsan

git clone https://github.com/pikarpov-LANL/Sapsan.git
cd Sapsan/
python setup.py install

If you experience any issues, you can try installing packages individually with:

pip install -r requirements.txt

Note: make sure you are using the latest release version

In order to create model graphs, Sapsan is using graphviz. If you would like to utilize this functionality, then please install graphviz via:

conda install graphviz

or

sudo apt-get install graphviz

In order to run Sapsan through Docker or build your own container to share, you will need to install it

pip install docker

Next, you can build a docker setup with the following:

make build-container

this will create a container named sapsan-docker.

If you want to run the container, type:

make run-container

a Jupyter notebook will be launched at localhost:7654

If you get the following error:

ImportError: libGL.so.1: cannot open shared object file: No such file or directory

your opencv-python package has some dependency issues. To resolve, try the following:

apt-get update
apt-get install ffmpeg libsm6 libxext6  -y