Computational experiments using entropy summary functions that appears in the paper "On the stability of persistent entropy and new summary functions for TDA".
It requires:
- Matlab with Javaplex installed (http://appliedtopology.github.io/javaplex/)
- R with the TDA package (https://cran.r-project.org/web/packages/TDA/index.html)
Functions used in NES example:
- Mosaicos6 is the main code of the experiment. It generates a
pattern, use it to create tesselations with different sizes
and then duplicate them adding noise.
Auxiliar functions:
- random_quad: generates a random non-degenerated quadilateral.
- quad_tilling: draw the tesselation and generate its point cloud.
- barcodes3: generate the barcodes.
- ent_summary3: calculate its NES function.
- Dist_L1: compute the L1 distance of a function.
Functions used in TES example:
- average_S1 is the main code of the experiment. It generates a list
feature.candidates[[i]][[j]] where i is the circle sample and j the
importance of the feature.
Auxiliar functions:
- Points_S1: generate a random sample of points in a circle.
- TES: Calculate the NES function and indicates the topological features.