Surface Volume Meshing Toolkit (SVMTK) is a Python frontend for specific 2D and 3D mesh generation methods in the Computational Geometry Algorithms Library (CGAL) .
SVMTK was designed to create volume meshes of soft organic tissue surfaces, like the human brain, with the option to repair unphysical errors on the surfaces.
First, ensure that all of the prerequisites needed by SVMTK are installed. Installation instructions for all of the SVMTK requirements can be found in REQUIREMENTS.md
Next, clone the SVMT repository by
git clone --recursive https://github.com/SVMTK/SVMTK
If you have already installed SVMTK you can update your installation with the following command executed in the /SVMTK directory
git pull origin master
git submodule update --init
Third, you will need to setup SVMTK if you have not already done so. To do this you can execute
python3 setup.py install
from within the SVMT directory. If you have, instead, updated an existing SVMT installation you should first clean any residual files by issuing the following command
python3 setup.py clean
It is advised, but not required, that you use a virtual environment for python. (see [https://docs.python.org/3/tutorial/venv.html] for instructions on using a python virtual environment)
You may see warnings as the setup process runs; these warnings arise from the CGAL source code, which is used by SVMTK, and are no cause for concern.
Finally, you can test your SVMTK installation by running
python3 setup.py test
Note that, if you do not use a virtual environment, you may need to preface the above commands, for setting up and testing SVMTK, with the `sudo' keyword. Thus,
sudo python3 setup.py install
sudo python3 setup.py clean
If the test is successful you should see the following output
----------------------------------------------------------------------
Ran 43 tests in XXXX s
OK
Also check the installation with any of the examples in examples/
There is a Dockerfile in the directory docker/.
For more information on how to use docker, take a look at the docker tutorial: [https://docs.docker.com/get-started/]
export PROJECT_HOME=path-to-base-install-dir
module load Python/3.7.4-GCCcore-8.3.0 CMake/3.12.1 GCC/8.3.0 Boost/1.71.0-GCC-8.3.0 Eigen/3.3.7 MPFR/4.0.2-GCCcore-8.3.0.lua GMP/6.1.2-GCCcore-8.3.0.lua
export CMAKE_PREFIX_PATH=$PROJECT_HOME/src/SVMTK/local:/cluster/software/Eigen/3.3.7:/cluster/software/Boost/1.71.0-GCC-8.3.0:/cluster/software/MPFR/4.0.2-GCCcore-8.3.0:/cluster/software/GMP/6.1.2-GCCcore-8.3.0
cd SVMTK/external/pybind11/
mkdir build
cd build/
cmake -DPYBIND11_TEST=off -DCMAKE_INSTALL_PREFIX=$PROJECT_HOME/src/SVMTK/local ..
make install
cd ../..
curl -sL https://github.com/CGAL/cgal/releases/download/releases%2FCGAL-5.0.3/CGAL-5.0.3.tar.xz | tar -xJf -
cd CGAL-5.0.3/
mkdir build
cd build/
cmake -DCMAKE_BUILD_TYPE=Release -DCMAKE_INSTALL_PREFIX=$PROJECT_HOME/src/SVMTK/local -DWITH_Eigen3:BOOL=ON .. && make -j4
make install
cd ../../../
mkdir -p $PROJECT_HOME/src/SVMTK/local/lib/python3.7/site-packages/
export PYTHONPATH=$PROJECT_HOME/src/SVMTK/local/lib/python3.7/site-packages:$PYTHONPATH
python setup.py install --prefix=$PROJECT_HOME/src/SVMTK/local
export PROJECT_HOME=path-to-base-install-dir
module load Python/3.7.4-GCCcore-8.3.0 CMake/3.12.1 GCC/8.3.0 Boost/1.71.0-GCC-8.3.0 Eigen/3.3.7 MPFR/4.0.2-GCCcore-8.3.0.lua GMP/6.1.2-GCCcore-8.3.0.lua
export PYTHONPATH=$PROJECT_HOME/src/SVMTK/local/lib/python3.7/site-packages:$PYTHONPATH
python -c "import SVMTK"
SVMTK is written by Lars Magnus Valnes and Jakob Schreiner