A suite of tools for cortical layer placement and analysis.
This package requires compiling some CGAL functions. You can install CGAL using any method you prefer; meson (which we use for building the package) just needs to locate CGAL with pkg-config. Here we describe the installation using Conan or Conda for managing the c++ dependencies.
First (using linux as example)
pip install conan
conan config install -tf profiles conan_profiles
conan install --profile:all linux --build=missing --output-folder=$PWD/conan_deps .
Then
pip install --config-settings=setup-args=--native-file=$PWD/conan_deps/conan_meson_native.ini --no-build-isolation --no-deps -e .
If you are using conda to manage build dependencies, then start by running the following script to generate the necessary pkg config files
python tools/generate_pkgs.py
This will generate pkg files for Boost, CGAL, Eigen, and TBB (TBB already exists but we want to include tbbmalloc in addition to tbb) and save them in $CONDA_PREFIX/lib/pkgconfig which is where conda stores its pkg config files (an alternative output directory can also be specified).
Note This path probably needs to be adjusted on windows!
You must also install pkg-config from conda as pkg-config provided by the system (e.g., in /bin/) will not search this particular directory
conda install pkg-config
Next, for an editable (developer) installation, use
pip install --no-build-isolation --no-deps -e .