Example application on a soundfile.
Performs the LLE alogrithm on a sound(file). As a first step, the sound is scanned piecewice with time windows sized t. In every scanning step, the time windows is shifted by dt. This generates an overall data set with dimension n=t and sample size N\approxt/dt. This dataset can then be embedded using LLE algorithm.
lle_sound(t = 500, dt = 20, k = 25, reg = 2, ss = FALSE,
p = 0.5, id = TRUE)t |
time window used to scan. |
dt |
time window shift. |
k |
number of neighbours. See |
reg |
regularisation method. See |
ss |
a logical values indicating wheather to perform subset selection. See |
p |
amount of data remaining after subset selection. See |
id |
a logical values indicating wheather to calculate the intrinsic dimension. See |
## Not run:
lle_sound()
lle_sound( 200, 20, 20 )
## End(Not run)Please choose more modern alternatives, such as Google Chrome or Mozilla Firefox.