Computes a set of n landmarks of a point cloud.
landmarks(x, n, dist_method = "euclidean", seed_index = 1, shuffle_data = FALSE)
| x | a data matrix. |
|---|---|
| n | the number of landmarks requested. |
| dist_method | the distance metric to use. Any distance measure in the |
| seed_index | the first landmark to seed the algorithm. |
| shuffle_data | whether to first randomly shuffle the data. |
This function uses the greedy maxmin procedure to produce a set of n evenly spaced landmark points in the data set
x. Maxmin is a simple greedy algorithm that is relatively efficient, but it has a tendency to pick out extremal points.
If the distance metric is euclidean, an efficient Rcpp implementation is used. If another metric is requested,
the algorithm is performed in R.
De Silva, Vin, and Gunnar E. Carlsson. "Topological estimation using witness complexes." SPBG 4 (2004): 157-166.