Home

The development philosophy of compGeometeR is to simply provide access in R to efficient computational geometry algorithms that have a wide range of applications. By keeping the functionality of compGeometeR at this lower level gives computational scientists maximum flexibility in how to use computational geometry in their work, and also makes maintenance of the package easier as the number of dependencies can be minimised.

By not providing bespoke functions that apply computational geometry algorithms to specific tasks does however mean that computational scientists will need to do some extra work, including using other packages, to make use of compGeometeR's functionality in their computational analyses. So to help computational scientists use compGeometeR these wiki pages provide a cookbook of code 'recipes' that show how to use computational geometry for different scientific questions. Computational scientists are free to learn from, adapt, and reuse these recipes in their own work - and we would welcome submissions of new recipes!

The compGeometeR cookbook

To try and make finding useful code recipes easier, the different ways compGeometeR computational geometry functionality can be applied have been broken into several categories:

Plotting and visualisation

• Plotting a 3D convex hull

Ecological niche modelling

• Convex hull fundamental niche model

High performance computing

• find_simplex parallelisation