R : Copyright 2003, The R Development Core Team
Version 1.6.2  (2003-01-10)

R is free software and comes with ABSOLUTELY NO WARRANTY.
You are welcome to redistribute it under certain conditions.
Type `license()' or `licence()' for distribution details.

R is a collaborative project with many contributors.
Type `contributors()' for more information.

Type `demo()' for some demos, `help()' for on-line help, or
`help.start()' for a HTML browser interface to help.
Type `q()' to quit R.

> invisible(options(echo = TRUE))
> library(nnet);
Loading required package: MASS 
> 
> train <- read.table("loans.dat",header=F,col.names=c("x1","x2","y"));
> 
> # Order is bias, wt, wt for input units to hidden units
> # followed by bias, wt, wt, wt, wt, wt for hidden units
> # to output units.
> 
> parms <- c(
+    23.896678, -62.450621,  11.170289,  
+    -3.845333,  -6.110258,  42.139228, 
+    -13.430940, -10.640723,  27.986805,  
+    21.322889, -10.303906, -77.733978,  
+    9.222935,  37.111371, -31.209375,  
+    12.395668,  33.377473, -15.457109, -19.592486,  51.958010, -28.515135);
> 
> fit <- nnet.formula(y ~ x1 + x2, data=train, size=5, maxit=500, Wts=parms);
# weights:  21
initial  value 8.072563 
iter  10 value 7.975028
iter  20 value 7.925210
iter  30 value 7.761261
iter  40 value 7.733799
iter  50 value 7.422430
iter  60 value 7.275243
iter  70 value 7.227319
iter  80 value 7.166369
iter  90 value 7.134806
iter 100 value 7.107312
iter 110 value 7.086445
iter 120 value 6.795647
iter 130 value 6.757341
iter 140 value 6.739297
iter 150 value 6.696500
iter 160 value 6.629398
iter 170 value 6.620332
iter 180 value 6.613965
iter 190 value 6.611035
iter 200 value 6.606162
iter 210 value 6.583114
iter 220 value 6.523906
iter 230 value 6.518200
iter 240 value 6.505579
iter 250 value 6.486478
iter 260 value 6.483777
iter 270 value 6.470495
iter 280 value 6.457123
iter 290 value 6.443233
iter 300 value 6.393421
iter 310 value 6.389902
iter 320 value 6.382983
iter 330 value 6.365094
iter 340 value 6.351842
iter 350 value 6.346412
iter 360 value 6.331224
iter 370 value 6.322801
iter 380 value 6.308089
iter 390 value 6.305907
iter 400 value 6.292051
iter 410 value 6.280831
iter 420 value 6.247631
iter 430 value 6.014635
iter 440 value 5.940287
iter 450 value 5.870574
iter 460 value 5.812439
iter 470 value 5.811569
iter 480 value 5.809648
iter 490 value 5.807886
iter 500 value 5.800171
final  value 5.800171 
stopped after 500 iterations
> 
> print(fit)
a 2-5-1 network with 21 weights
inputs: x1 x2 
output(s): y 
options were -
> print.summary.nnet(fit)
a 2-5-1 network with 21 weights
options were -
  b->h1  i1->h1  i2->h1 
 142.65 -416.24   81.16 
  b->h2  i1->h2  i2->h2 
  18.87   -3.11   51.81 
  b->h3  i1->h3  i2->h3 
 -89.74  -38.27  180.18 
  b->h4  i1->h4  i2->h4 
  12.61   -7.68  -40.95 
  b->h5  i1->h5  i2->h5 
  -1.78   94.57  -45.50 
   b->o   h1->o   h2->o   h3->o   h4->o   h5->o 
  29.65   50.19    0.94  -73.24   92.44  -80.44 
> 
> size <- 100;
> 
> grid <- mat.or.vec((size+1)*(size+1),2);
> for (i in 0:size) {
+ for (j in 0:size) {
+   ij <- (size+1)*j+i+1
+   grid[ij,1] <- i;
+   grid[ij,2] <- j;
+ }
+ }
> grid <- grid/(size+1);
> 
> Grid <- data.frame(x1=grid[,1],x2=grid[,2]);
> 
> dflt <- predict.nnet(object=fit,newdat=Grid);
> dflt <- (dflt > 0.5);
> 
> edge <- mat.or.vec((size+1)*(size+1),1);
> edge <- as.logical(edge);
> for (i in 1:size) {
+ for (j in 1:size) {
+   left <- (size+1)*j+i+1;
+   rite <- (size+1)*(j-1)+i+1;
+   top  <- left;
+   bot  <- left - 1;
+   edge[left] <- 
+     ( dflt[left] && !dflt[rite] ) ||
+     ( !dflt[left] && dflt[rite] ) ||
+     ( dflt[top]  && !dflt[bot]  ) || 
+     ( !dflt[top]  && dflt[bot]  ) ;
+ }
+ }
> 
> X <- cbind(train$x1,train$x2);
> y <- (train$y==1);
> 
> source("psopts.r");
> postscript(file="nnet0.eps");
> 
> plot(X,type='n',xlab="FICO Score",ylab="P-Index");
> points(X[y,],pch='o',col="red");
> points(X[!y,],pch='o',col="green");
> points(grid[dflt,],pch='.',col="red");
> points(grid[!dflt,],pch='.',col="green");
> points(grid[edge,],pch='+',col="black",cex=0.4);
> 
> dev.off();
null device 
          1 
> 
> test <- read.table("eval.dat",header=F,col.names=c("x1","x2","y"));
> 
> dflt <- predict.nnet(object=fit,newdat=test);
> dflt <- (dflt > 0.5);
> 
> dflt <- as.numeric(dflt);
> ty <- test$y;
> 
> err <- abs(ty-dflt);
> print(mean(err));
[1] 0.09243
> 
> proc.time()
[1] 4.57 0.22 4.92 0.00 0.00
>