bucket: Variable averaging (bucketing) for data matrices

In wccsom: SOM Networks for Comparing Patterns with Peak Shifts

Description

Function bucket decreases the size (i.e., the number of columns) of a data matrix by averaging variables. Function debucket achieves the reverse by linear interpolation.

Usage

bucket(x, factor)
debucket(x, nout)

Arguments

x	Data matrix: each variable corresponds with a column.
factor	Bucket factor: this number of variables will be averaged.
nout	Required number of variables after debucketing.

Value

Returns a data matrix of the new dimensions.

Author(s)

Ron Wehrens

Examples

data(cepha)
gr <- somgrid(3,3, "hexagonal")

X <- bucket(cepha$patterns, 4)
system.time(x <- wccsom(X, grid=gr, trwidth=5,
                        rlen=500, FineTune=FALSE))
## this is much faster than the following code:
## Not run: 
set.seed(7)
system.time(x <- wccsom(cepha$patterns, grid=gr, trwidth=20,
                        rlen=500, FineTune=FALSE))

## End(Not run)

Example output

Loading required package: class
Loading required package: MASS
Loading required package: kohonen

Attaching package: 'kohonen'

The following object is masked from 'package:class':

    somgrid


Attaching package: 'wccsom'

The following object is masked from 'package:kohonen':

    unit.distances

0%....20%....40%....60%....80%....
   user  system elapsed 
  0.128   0.000   0.129 
0%....20%....40%....60%....80%....
   user  system elapsed 
  2.148   0.007   2.172 

wccsom documentation built on May 30, 2017, 12:52 a.m.