lle_rectangular
(Package: lle) :
Example on a rectangular signal.
A rectangular signal with variable width and distance is generated and analysed (see figure). Every sample has dimension n=500. In the example N samples are generated, that only differ by the two parameters width and distance. As a result it can be seen that the data can be embedded from a 500- into a 2-dimensional space which's limits are specified by the limits of the two parameters.
● Data Source:
CranContrib
● Keywords:
● Alias: lle_rectangular
●
0 images
|
lle
(Package: lle) :
Locally linear embedding main function.
Performs all steps of LLE algorithm by calling the functions of the package.
● Data Source:
CranContrib
● Keywords:
● Alias: lle
●
0 images
|
lle_swissrole
(Package: lle) :
Example application on a swissrole.
Like the S-curve this is a standard example to demonstrate the functioning of LLE.
● Data Source:
CranContrib
● Keywords:
● Alias: lle_swissrole
●
0 images
|
find_nn_eps
(Package: lle) :
Find nearest neighbours in epsilon environment.
Finds the nearest points for every point in an environment with radius eps .
● Data Source:
CranContrib
● Keywords:
● Alias: find_nn_eps
●
0 images
|
lle_scurve
(Package: lle) :
Example application on a three-dimensional S-curve dataset.
Performs the LLE alogrithm on a three dimensional S-curve, which is a standard example for data embedding algorithms.
● Data Source:
CranContrib
● Keywords:
● Alias: lle_scurve
●
0 images
|
find_weights
(Package: lle) :
Calculate weight matrix.
Calculates the weights for every neighbour of x_i.
● Data Source:
CranContrib
● Keywords:
● Alias: find_weights
●
0 images
|
lle_spiral
(Package: lle) :
Example application on a n-dimensional spiral
In this example nine three-dimensional spirals are generated. Every spiral has a higher number of windings than the previous. The first spiral's intrinsic dimension is determined as m=1. It can be observed that a higher number of windings with constant sampling rate leads to a higher calculated intrinsic dimension. Due to this fact, the last spiral's intrinsic dimension is computed as m=2.
● Data Source:
CranContrib
● Keywords:
● Alias: lle_spiral
●
0 images
|
calc_k
(Package: lle) :
Calculation of the optimal number of neighbours.
Calculates of optimal number of neighbours by using the algorithm proposed by Kayo (see Ref.). Therefore the LLE alorithm is performed for every k-neighbourhood size.
● Data Source:
CranContrib
● Keywords:
● Alias: calc_k
●
0 images
|
lle_wave
(Package: lle) :
wavefile
syntheticly generated wavefile with 10 modes.
● Data Source:
CranContrib
● Keywords:
● Alias: lle_wave
●
0 images
|
plot_lle
(Package: lle) :
Plot LLE results
Function for plotting LLE results either in static or in dynamic way.
● Data Source:
CranContrib
● Keywords:
● Alias: plot_lle
●
0 images
|