Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Library: rpclib
See also: xgobilocpol xgobilowess xgobinadawat

Quantlet: xgobineunet
Description: conmputes a neural network fit for XGobi

Link:
Usage: yh = xgobineunet(x,y,g,hidden)
Input:
x n x 1 matrix
y n x 1 matrix
g n x 1 matrix with integers starting at zero
hidden scalar number of hidden units
Output:
yh n x 1 matrix smoothed fit

Note:

Example:

; load the library for XGobi

library ("rpclib")

; create a display

d=createdisplay(1,1)

; read a data set

x=read("motcyc")

; generate randomly three subgroups

g=floor(3.*uniform(rows(x)))

; color the data after the subgroups

setmaskp (x, g, 3, 4)

; show the data and the smooth

show (d, 1, 1, x, x[,1]~xgobineunet(x[,1],x[,2],g,10))

Result:

shows a neural network fit of for the three 

different subgroups in the data


Library: rpclib
See also: xgobilocpol xgobilowess xgobinadawat

Keywords - Function groups - @ A B C D E F G H I J K L M N O P Q R S T U V W X Y Z

Author: Sigbert Klinke, 960709
(C) MD*TECH Method and Data Technologies, 21.9.2000