#Instalación de paquetes y llamados de librerias
#install.packages("neuralnet")
library(neuralnet)
#Datos
set.seed(42) # Establece la semilla aleatoria para asegurar reproducibilidad
data <- data.frame(
Cat1 = round(runif(100, min = 1, max = 10)),
Cat2 = round(runif(100, min = 1, max = 10)))
head(data)
#Columna adicional (data frame)
data$Hyp <- sqrt(data$Cat1*data$Cat1 + data$Cat2*data$Cat2)
head(data)
#Validación
fold.test <- sample(nrow(data), nrow(data) / 3)
fold.test
## [1] 24 98 25 92 61 62 14 34 66 32 27 10 57 28 37 89 5 35 78 94 87 54 31 43 52
## [26] 59 90 30 76 17 95 13 63
test <- data[fold.test, ]
train <- data[-fold.test, ]
head(test)
head(train)
#Entrenamiento red neuronal
ann <- neuralnet(Hyp ~ Cat1 + Cat2, train, hidden = 10, rep = 3)
ann
## $call
## neuralnet(formula = Hyp ~ Cat1 + Cat2, data = train, hidden = 10,
## rep = 3)
##
## $response
## Hyp
## 1 11.401754
## 2 9.486833
## 3 5.000000
## 4 9.433981
## 6 11.661904
## 7 11.313708
## 8 8.246211
## 9 9.219544
## 11 7.810250
## 12 11.401754
## 15 7.810250
## 16 10.816654
## 18 4.472136
## 19 8.602325
## 20 10.000000
## 21 9.848858
## 22 5.385165
## 23 11.661904
## 26 7.810250
## 29 5.830952
## 33 8.062258
## 36 10.000000
## 38 3.605551
## 39 12.041595
## 40 9.899495
## 41 4.472136
## 42 5.385165
## 44 12.806248
## 45 9.433981
## 46 12.806248
## 47 9.486833
## 48 12.206556
## 49 10.770330
## 50 7.280110
## 51 8.062258
## 53 9.433981
## 55 7.071068
## 56 11.313708
## 58 3.162278
## 60 9.219544
## 64 6.708204
## 65 10.816654
## 67 5.830952
## 68 8.944272
## 69 7.280110
## 70 4.242641
## 71 8.062258
## 72 5.385165
## 73 5.830952
## 74 7.071068
## 75 5.830952
## 77 8.062258
## 79 10.000000
## 80 8.062258
## 81 10.816654
## 82 9.219544
## 83 5.656854
## 84 7.615773
## 85 11.313708
## 86 10.000000
## 88 8.246211
## 91 7.615773
## 93 3.605551
## 96 8.062258
## 97 7.211103
## 99 11.313708
## 100 10.630146
##
## $covariate
## Cat1 Cat2
## 1 9 7
## 2 9 3
## 3 4 3
## 4 8 5
## 6 6 10
## 7 8 8
## 8 2 8
## 9 7 6
## 11 5 6
## 12 7 9
## 15 5 6
## 16 9 6
## 18 2 4
## 19 5 7
## 20 6 8
## 21 9 4
## 22 2 5
## 23 10 6
## 26 6 5
## 29 5 3
## 33 4 7
## 36 8 6
## 38 3 2
## 39 9 8
## 40 7 7
## 41 4 2
## 42 5 2
## 44 10 8
## 45 5 8
## 46 10 8
## 47 9 3
## 48 7 10
## 49 10 4
## 50 7 2
## 51 4 7
## 53 5 8
## 55 1 7
## 56 8 8
## 58 3 1
## 60 6 7
## 64 6 3
## 65 9 6
## 67 3 5
## 68 8 4
## 69 7 2
## 70 3 3
## 71 1 8
## 72 2 5
## 73 3 5
## 74 5 5
## 75 3 5
## 77 1 8
## 79 6 8
## 80 1 8
## 81 6 9
## 82 2 9
## 83 4 4
## 84 7 3
## 85 8 8
## 86 6 8
## 88 2 8
## 91 7 3
## 93 3 2
## 96 8 1
## 97 4 6
## 99 8 8
## 100 7 8
##
## $model.list
## $model.list$response
## [1] "Hyp"
##
## $model.list$variables
## [1] "Cat1" "Cat2"
##
##
## $err.fct
## function (x, y)
## {
## 1/2 * (y - x)^2
## }
## <bytecode: 0x000001cc2d3434e0>
## <environment: 0x000001cc2d347d88>
## attr(,"type")
## [1] "sse"
##
## $act.fct
## function (x)
## {
## 1/(1 + exp(-x))
## }
## <bytecode: 0x000001cc2d33a978>
## <environment: 0x000001cc2d33a080>
## attr(,"type")
## [1] "logistic"
##
## $linear.output
## [1] TRUE
##
## $data
## Cat1 Cat2 Hyp
## 1 9 7 11.401754
## 2 9 3 9.486833
## 3 4 3 5.000000
## 4 8 5 9.433981
## 6 6 10 11.661904
## 7 8 8 11.313708
## 8 2 8 8.246211
## 9 7 6 9.219544
## 11 5 6 7.810250
## 12 7 9 11.401754
## 15 5 6 7.810250
## 16 9 6 10.816654
## 18 2 4 4.472136
## 19 5 7 8.602325
## 20 6 8 10.000000
## 21 9 4 9.848858
## 22 2 5 5.385165
## 23 10 6 11.661904
## 26 6 5 7.810250
## 29 5 3 5.830952
## 33 4 7 8.062258
## 36 8 6 10.000000
## 38 3 2 3.605551
## 39 9 8 12.041595
## 40 7 7 9.899495
## 41 4 2 4.472136
## 42 5 2 5.385165
## 44 10 8 12.806248
## 45 5 8 9.433981
## 46 10 8 12.806248
## 47 9 3 9.486833
## 48 7 10 12.206556
## 49 10 4 10.770330
## 50 7 2 7.280110
## 51 4 7 8.062258
## 53 5 8 9.433981
## 55 1 7 7.071068
## 56 8 8 11.313708
## 58 3 1 3.162278
## 60 6 7 9.219544
## 64 6 3 6.708204
## 65 9 6 10.816654
## 67 3 5 5.830952
## 68 8 4 8.944272
## 69 7 2 7.280110
## 70 3 3 4.242641
## 71 1 8 8.062258
## 72 2 5 5.385165
## 73 3 5 5.830952
## 74 5 5 7.071068
## 75 3 5 5.830952
## 77 1 8 8.062258
## 79 6 8 10.000000
## 80 1 8 8.062258
## 81 6 9 10.816654
## 82 2 9 9.219544
## 83 4 4 5.656854
## 84 7 3 7.615773
## 85 8 8 11.313708
## 86 6 8 10.000000
## 88 2 8 8.246211
## 91 7 3 7.615773
## 93 3 2 3.605551
## 96 8 1 8.062258
## 97 4 6 7.211103
## 99 8 8 11.313708
## 100 7 8 10.630146
##
## $exclude
## NULL
##
## $net.result
## $net.result[[1]]
## [,1]
## 1 11.402612
## 2 9.519214
## 3 4.984103
## 4 9.405900
## 6 11.656256
## 7 11.322846
## 8 8.244788
## 9 9.231958
## 11 7.804752
## 12 11.416441
## 15 7.804752
## 16 10.826106
## 18 4.492108
## 19 8.587124
## 20 9.987587
## 21 9.858982
## 22 5.369898
## 23 11.687492
## 26 7.819815
## 29 5.832493
## 33 8.069916
## 36 9.994650
## 38 3.623441
## 39 12.030252
## 40 9.899944
## 41 4.464623
## 42 5.393795
## 44 12.796030
## 45 9.446688
## 46 12.796030
## 47 9.519214
## 48 12.195183
## 49 10.718863
## 50 7.292632
## 51 8.069916
## 53 9.446688
## 55 7.070216
## 56 11.322846
## 58 3.140956
## 60 9.199147
## 64 6.713193
## 65 10.826106
## 67 5.837452
## 68 8.922016
## 69 7.292632
## 70 4.239743
## 71 8.063106
## 72 5.369898
## 73 5.837452
## 74 7.073729
## 75 5.837452
## 77 8.063106
## 79 9.987587
## 80 8.063106
## 81 10.835795
## 82 9.215384
## 83 5.646613
## 84 7.607442
## 85 11.322846
## 86 9.987587
## 88 8.244788
## 91 7.607442
## 93 3.623441
## 96 8.037032
## 97 7.215242
## 99 11.322846
## 100 10.627308
##
## $net.result[[2]]
## [,1]
## 1 11.399931
## 2 9.492859
## 3 4.999489
## 4 9.436373
## 6 11.632796
## 7 11.318072
## 8 8.269215
## 9 9.218777
## 11 7.808101
## 12 11.420286
## 15 7.808101
## 16 10.824492
## 18 4.448902
## 19 8.598999
## 20 9.993750
## 21 9.850392
## 22 5.378203
## 23 11.667929
## 26 7.802915
## 29 5.824440
## 33 8.071776
## 36 10.003219
## 38 3.633644
## 39 12.042164
## 40 9.896263
## 41 4.466560
## 42 5.370484
## 44 12.796950
## 45 9.425151
## 46 12.796950
## 47 9.492859
## 48 12.227657
## 49 10.752988
## 50 7.283864
## 51 8.071776
## 53 9.425151
## 55 7.053032
## 56 11.318072
## 58 3.136049
## 60 9.213628
## 64 6.704215
## 65 10.824492
## 67 5.840270
## 68 8.943130
## 69 7.283864
## 70 4.254772
## 71 8.056886
## 72 5.378203
## 73 5.840270
## 74 7.063123
## 75 5.840270
## 77 8.056886
## 79 9.993750
## 80 8.056886
## 81 10.812661
## 82 9.218738
## 83 5.650183
## 84 7.616446
## 85 11.318072
## 86 9.993750
## 88 8.269215
## 91 7.616446
## 93 3.633644
## 96 8.060178
## 97 7.220886
## 99 11.318072
## 100 10.631394
##
## $net.result[[3]]
## [,1]
## 1 11.412775
## 2 9.494696
## 3 5.013533
## 4 9.435911
## 6 11.642046
## 7 11.306745
## 8 8.247518
## 9 9.214782
## 11 7.796409
## 12 11.407147
## 15 7.796409
## 16 10.825445
## 18 4.452021
## 19 8.597881
## 20 10.009269
## 21 9.842302
## 22 5.379431
## 23 11.661073
## 26 7.807053
## 29 5.838968
## 33 8.061572
## 36 10.003654
## 38 3.623500
## 39 12.040854
## 40 9.892191
## 41 4.449067
## 42 5.352455
## 44 12.812331
## 45 9.446007
## 46 12.812331
## 47 9.494696
## 48 12.194443
## 49 10.739118
## 50 7.279692
## 51 8.061572
## 53 9.446007
## 55 7.071937
## 56 11.306745
## 58 3.150028
## 60 9.210534
## 64 6.713102
## 65 10.825445
## 67 5.835899
## 68 8.944429
## 69 7.279692
## 70 4.259053
## 71 8.062017
## 72 5.379431
## 73 5.835899
## 74 7.066216
## 75 5.835899
## 77 8.062017
## 79 10.009269
## 80 8.062017
## 81 10.837696
## 82 9.218406
## 83 5.669590
## 84 7.620836
## 85 11.306745
## 86 10.009269
## 88 8.247518
## 91 7.620836
## 93 3.623500
## 96 8.072744
## 97 7.205223
## 99 11.306745
## 100 10.627805
##
##
## $weights
## $weights[[1]]
## $weights[[1]][[1]]
## [,1] [,2] [,3] [,4] [,5] [,6] [,7]
## [1,] -3.4736901 5.4073092 3.1066176 -0.6348233 4.784375 -4.8164186 0.7798317
## [2,] 0.6583925 -0.3948012 0.2357640 0.2498918 8.438418 -0.1267209 -0.1590813
## [3,] -0.3237344 -0.2354862 -0.9629211 0.1946966 6.644521 0.6328688 -0.2048647
## [,8] [,9] [,10]
## [1,] -1.3334280 2.8756784 6.0829306
## [2,] -1.7961632 -0.2936289 -1.1953931
## [3,] -0.2292374 -0.1467669 0.5471151
##
## $weights[[1]][[2]]
## [,1]
## [1,] 1.629905
## [2,] 4.883070
## [3,] -2.505219
## [4,] -1.135234
## [5,] 6.617429
## [6,] 2.073334
## [7,] 3.921345
## [8,] -3.244486
## [9,] 34.696889
## [10,] -1.326125
## [11,] 1.307099
##
##
## $weights[[2]]
## $weights[[2]][[1]]
## [,1] [,2] [,3] [,4] [,5] [,6]
## [1,] -2.49308821 -4.85940059 5.6836119 -0.80252968 -0.86330719 -1.7722186
## [2,] -0.02234665 -0.09449356 -0.6067413 0.07397678 0.18925057 0.3892254
## [3,] 0.56724582 0.53640461 0.1540598 0.19065416 -0.04987882 -0.1056768
## [,7] [,8] [,9] [,10]
## [1,] -1.1985103 -0.3287892 32.1245331 -2.1898697
## [2,] 0.3111842 -0.1422862 0.3275825 -0.2883928
## [3,] 0.4985407 -0.1660481 0.3341107 0.5428211
##
## $weights[[2]][[2]]
## [,1]
## [1,] -0.02610686
## [2,] 3.16110910
## [3,] 5.64224822
## [4,] -3.26792076
## [5,] 0.82600052
## [6,] 1.31590951
## [7,] 6.79040267
## [8,] 1.97345626
## [9,] -1.97855072
## [10,] 2.17045225
## [11,] 1.49122726
##
##
## $weights[[3]]
## $weights[[3]][[1]]
## [,1] [,2] [,3] [,4] [,5] [,6]
## [1,] 2.2715080 -2.69311672 0.3099679 0.89272080 -2.6217216 -1.704533270
## [2,] 0.6468456 0.00786083 -4.5785888 -0.04944209 0.4982159 0.390716471
## [3,] -0.6896373 0.29902722 0.3421396 -0.35774376 0.6349624 0.008370674
## [,7] [,8] [,9] [,10]
## [1,] -0.08293560 -4.32964779 -4.5696642 -2.43389864
## [2,] 0.09799909 0.42321907 -1.5446363 0.39089759
## [3,] -0.72118946 -0.04149045 0.8238176 0.06987496
##
## $weights[[3]][[2]]
## [,1]
## [1,] 1.958606
## [2,] -1.001807
## [3,] 7.473648
## [4,] -1.055418
## [5,] -2.802842
## [6,] 1.724415
## [7,] 2.212220
## [8,] 2.022351
## [9,] 5.763455
## [10,] 1.152008
## [11,] 3.050540
##
##
##
## $generalized.weights
## $generalized.weights[[1]]
## [,1] [,2]
## 1 -0.0068387734 -0.005028759
## 2 -0.0111181689 -0.003792956
## 3 -0.0405733387 -0.029739217
## 4 -0.0107899724 -0.006926797
## 6 -0.0040235456 -0.006226108
## 7 -0.0059886111 -0.006176084
## 8 -0.0051445161 -0.016848904
## 9 -0.0099121257 -0.008569274
## 11 -0.0119774363 -0.014139896
## 12 -0.0052532686 -0.006731379
## 15 -0.0119774363 -0.014139896
## 16 -0.0081783562 -0.005250853
## 18 -0.0245577337 -0.054828011
## 19 -0.0086129574 -0.012586796
## 20 -0.0065934516 -0.009229358
## 21 -0.0105099060 -0.004375356
## 22 -0.0160146365 -0.037911215
## 23 -0.0065979214 -0.004369855
## 26 -0.0144696829 -0.012156195
## 29 -0.0311485901 -0.017595017
## 33 -0.0083861364 -0.015606483
## 36 -0.0087450162 -0.006877152
## 38 -0.0802778426 -0.057109404
## 39 -0.0054945434 -0.004996862
## 40 -0.0082495976 -0.007848644
## 41 -0.0583280017 -0.029437794
## 42 -0.0396166961 -0.016501014
## 44 -0.0052373806 -0.003835245
## 45 -0.0061778246 -0.011187731
## 46 -0.0052373806 -0.003835245
## 47 -0.0111181689 -0.003792956
## 48 -0.0042343842 -0.005423739
## 49 -0.0075987472 -0.003824173
## 50 -0.0217701467 -0.006053637
## 51 -0.0083861364 -0.015606483
## 53 -0.0061778246 -0.011187731
## 55 0.0027319215 -0.022874176
## 56 -0.0059886111 -0.006176084
## 58 -0.1241273717 -0.063879815
## 60 -0.0087862373 -0.009888721
## 64 -0.0228905439 -0.011350009
## 65 -0.0081783562 -0.005250853
## 67 -0.0187707580 -0.030234407
## 68 -0.0131592103 -0.005868123
## 69 -0.0217701467 -0.006053637
## 70 -0.0490982861 -0.050422420
## 71 0.0007393125 -0.017323882
## 72 -0.0160146365 -0.037911215
## 73 -0.0187707580 -0.030234407
## 74 -0.0165177969 -0.016564437
## 75 -0.0187707580 -0.030234407
## 77 0.0007393125 -0.017323882
## 79 -0.0065934516 -0.009229358
## 80 0.0007393125 -0.017323882
## 81 -0.0050070795 -0.007992820
## 82 -0.0040452293 -0.012125029
## 83 -0.0267241101 -0.027690747
## 84 -0.0184379788 -0.007212471
## 85 -0.0059886111 -0.006176084
## 86 -0.0065934516 -0.009229358
## 88 -0.0051445161 -0.016848904
## 91 -0.0184379788 -0.007212471
## 93 -0.0802778426 -0.057109404
## 96 -0.0168877586 -0.004722078
## 97 -0.0121861142 -0.018360371
## 99 -0.0059886111 -0.006176084
## 100 -0.0066455830 -0.007472418
##
## $generalized.weights[[2]]
## [,1] [,2]
## 1 -0.006655036 -0.005104975
## 2 -0.011671075 -0.003773778
## 3 -0.039444617 -0.029595542
## 4 -0.010720036 -0.006713125
## 6 -0.004537590 -0.006457451
## 7 -0.006049442 -0.006210841
## 8 -0.004143678 -0.015970487
## 9 -0.010074640 -0.008573876
## 11 -0.011912732 -0.014504760
## 12 -0.005338803 -0.006785861
## 15 -0.011912732 -0.014504760
## 16 -0.007865561 -0.005155075
## 18 -0.032605181 -0.058277826
## 19 -0.008779024 -0.012385287
## 20 -0.006750701 -0.008929859
## 21 -0.010460182 -0.004692986
## 22 -0.017590238 -0.040545504
## 23 -0.006775193 -0.004096329
## 26 -0.014541772 -0.012145312
## 29 -0.030489708 -0.018126772
## 33 -0.008392010 -0.015109486
## 36 -0.008928723 -0.006623235
## 38 -0.082373839 -0.058293539
## 39 -0.005572785 -0.005131940
## 40 -0.008062755 -0.008015097
## 41 -0.056398818 -0.030764852
## 42 -0.039604513 -0.017129337
## 44 -0.005067754 -0.004223489
## 45 -0.006637936 -0.010601075
## 46 -0.005067754 -0.004223489
## 47 -0.011671075 -0.003773778
## 48 -0.004546800 -0.005787357
## 49 -0.008420327 -0.003692816
## 50 -0.021277544 -0.006211337
## 51 -0.008392010 -0.015109486
## 53 -0.006637936 -0.010601075
## 55 -0.005144431 -0.023657748
## 56 -0.006049442 -0.006210841
## 58 -0.128352981 -0.065579227
## 60 -0.008617893 -0.009998821
## 64 -0.023505588 -0.011527755
## 65 -0.007865561 -0.005155075
## 67 -0.018104151 -0.030927093
## 68 -0.012651153 -0.006294605
## 69 -0.021277544 -0.006211337
## 70 -0.050382972 -0.049444685
## 71 -0.003141686 -0.017523391
## 72 -0.017590238 -0.040545504
## 73 -0.018104151 -0.030927093
## 74 -0.016430798 -0.016673851
## 75 -0.018104151 -0.030927093
## 77 -0.003141686 -0.017523391
## 79 -0.006750701 -0.008929859
## 80 -0.003141686 -0.017523391
## 81 -0.005428455 -0.007813961
## 82 -0.002884258 -0.012290046
## 83 -0.026606261 -0.026906428
## 84 -0.018330275 -0.007639597
## 85 -0.006049442 -0.006210841
## 86 -0.006750701 -0.008929859
## 88 -0.004143678 -0.015970487
## 91 -0.018330275 -0.007639597
## 93 -0.082373839 -0.058293539
## 96 -0.017956338 -0.003026447
## 97 -0.011996696 -0.018651303
## 99 -0.006049442 -0.006210841
## 100 -0.006494416 -0.007466318
##
## $generalized.weights[[3]]
## [,1] [,2]
## 1 -0.0067162593 -0.005132269
## 2 -0.0116904985 -0.003572434
## 3 -0.0394750129 -0.030516157
## 4 -0.0106592124 -0.006696912
## 6 -0.0044721959 -0.006245190
## 7 -0.0060784621 -0.005986618
## 8 -0.0018801922 -0.016208073
## 9 -0.0101428612 -0.008535156
## 11 -0.0120225919 -0.014489325
## 12 -0.0049578741 -0.006666525
## 15 -0.0120225919 -0.014489325
## 16 -0.0078296265 -0.005285191
## 18 -0.0325387066 -0.058447283
## 19 -0.0088035164 -0.012724385
## 20 -0.0065357324 -0.009106327
## 21 -0.0103611104 -0.004644613
## 22 -0.0169204103 -0.040205463
## 23 -0.0067020091 -0.004291474
## 26 -0.0145150888 -0.011835403
## 29 -0.0301848069 -0.018804838
## 33 -0.0087312862 -0.015305054
## 36 -0.0089681251 -0.006657193
## 38 -0.0814193748 -0.059023362
## 39 -0.0056938794 -0.004846971
## 40 -0.0080970587 -0.008046433
## 41 -0.0567289789 -0.033266592
## 42 -0.0399651952 -0.018952912
## 44 -0.0051573966 -0.004002287
## 45 -0.0067518837 -0.010735051
## 46 -0.0051573966 -0.004002287
## 47 -0.0116904985 -0.003572434
## 48 -0.0040589523 -0.005666529
## 49 -0.0084749311 -0.003534702
## 50 -0.0215701139 -0.006414507
## 51 -0.0087312862 -0.015305054
## 53 -0.0067518837 -0.010735051
## 55 -0.0119460892 -0.023832306
## 56 -0.0060784621 -0.005986618
## 58 -0.1157257162 -0.056717814
## 60 -0.0085815384 -0.010196636
## 64 -0.0232819117 -0.011929660
## 65 -0.0078296265 -0.005285191
## 67 -0.0181698307 -0.030338310
## 68 -0.0125363284 -0.006274885
## 69 -0.0215701139 -0.006414507
## 70 -0.0511225227 -0.050506450
## 71 -0.0118998525 -0.016607646
## 72 -0.0169204103 -0.040205463
## 73 -0.0181698307 -0.030338310
## 74 -0.0164805226 -0.016145600
## 75 -0.0181698307 -0.030338310
## 77 -0.0118998525 -0.016607646
## 79 -0.0065357324 -0.009106327
## 80 -0.0118998525 -0.016607646
## 81 -0.0052016640 -0.007748979
## 82 -0.0005100559 -0.012734801
## 83 -0.0266391807 -0.026273093
## 84 -0.0182520681 -0.007786219
## 85 -0.0060784621 -0.005986618
## 86 -0.0065357324 -0.009106327
## 88 -0.0018801922 -0.016208073
## 91 -0.0182520681 -0.007786219
## 93 -0.0814193748 -0.059023362
## 96 -0.0184669575 -0.003203625
## 97 -0.0121479043 -0.018665219
## 99 -0.0060784621 -0.005986618
## 100 -0.0063394459 -0.007440239
##
##
## $startweights
## $startweights[[1]]
## $startweights[[1]][[1]]
## [,1] [,2] [,3] [,4] [,5] [,6]
## [1,] 0.09992556 1.604873 1.2094427 -0.4214492 -1.6459002 -2.0816460
## [2,] 0.30272834 -1.102002 0.7903881 -0.2435677 2.1120614 -1.8857641
## [3,] -1.15864442 -0.823719 -1.0996495 -0.3374156 -0.9502351 0.6179856
## [,7] [,8] [,9] [,10]
## [1,] -0.2885732 -1.8450177 1.4803111 2.040190
## [2,] -1.3421235 -0.1393937 -0.9992088 -1.322056
## [3,] 0.2610382 -0.4468103 -0.5044699 1.479531
##
## $startweights[[1]][[2]]
## [,1]
## [1,] 0.39402377
## [2,] 2.35729914
## [3,] -0.26948844
## [4,] -0.58341584
## [5,] 1.51520303
## [6,] 1.00274142
## [7,] 0.67680762
## [8,] -2.82309722
## [9,] -0.25522942
## [10,] -0.04921993
## [11,] -0.47624432
##
##
## $startweights[[2]]
## $startweights[[2]][[1]]
## [,1] [,2] [,3] [,4] [,5] [,6]
## [1,] -0.5469122 -1.7382164 0.19253010 1.65972336 2.0518441 -0.5594724
## [2,] -1.3351214 -0.3230168 -0.97561601 -0.07401147 0.2084942 -0.4620188
## [3,] -0.3590329 1.3411185 -0.06092112 0.39210617 1.4405233 -1.2568005
## [,7] [,8] [,9] [,10]
## [1,] 1.8313707 -0.07503766 1.79520556 -0.74043062
## [2,] -1.2346057 -0.58247131 -1.36527440 -0.02318738
## [3,] 0.9530732 0.90992743 0.06218088 1.32215953
##
## $startweights[[2]][[2]]
## [,1]
## [1,] -1.43531206
## [2,] 0.03319267
## [3,] -0.20503696
## [4,] -1.23730425
## [5,] -0.62810790
## [6,] -0.29000247
## [7,] 0.20587115
## [8,] 0.58845903
## [9,] -1.02433401
## [10,] 0.76507400
## [11,] -1.55436796
##
##
## $startweights[[3]]
## $startweights[[3]][[1]]
## [,1] [,2] [,3] [,4] [,5] [,6]
## [1,] -0.04323551 -0.41272065 -1.3794058 -0.04281657 -0.0861781 -0.2614958
## [2,] 0.88349955 0.03549472 -2.1333457 1.41407308 0.8695214 0.4106070
## [3,] -1.45838599 0.11433620 -0.7488424 1.05871406 -1.5386820 -0.5045991
## [,7] [,8] [,9] [,10]
## [1,] 0.4470158 -1.4545869 1.089467 -0.09123807
## [2,] 1.0183163 0.3584167 -2.981564 1.31682260
## [3,] -1.4114870 -0.8825137 1.661799 0.79221066
##
## $startweights[[3]][[2]]
## [,1]
## [1,] 0.1943478
## [2,] -1.1160341
## [3,] -0.1253143
## [4,] -0.6790750
## [5,] -1.6581750
## [6,] -0.0396960
## [7,] -0.2113846
## [8,] -0.9029063
## [9,] -0.5096644
## [10,] -1.8410224
## [11,] -0.9263333
##
##
##
## $result.matrix
## [,1] [,2] [,3]
## error 7.010750e-03 4.178217e-03 3.854178e-03
## reached.threshold 9.629892e-03 9.915530e-03 9.645923e-03
## steps 1.795300e+04 3.394100e+04 1.812600e+04
## Intercept.to.1layhid1 -3.473690e+00 -2.493088e+00 2.271508e+00
## Cat1.to.1layhid1 6.583925e-01 -2.234665e-02 6.468456e-01
## Cat2.to.1layhid1 -3.237344e-01 5.672458e-01 -6.896373e-01
## Intercept.to.1layhid2 5.407309e+00 -4.859401e+00 -2.693117e+00
## Cat1.to.1layhid2 -3.948012e-01 -9.449356e-02 7.860830e-03
## Cat2.to.1layhid2 -2.354862e-01 5.364046e-01 2.990272e-01
## Intercept.to.1layhid3 3.106618e+00 5.683612e+00 3.099679e-01
## Cat1.to.1layhid3 2.357640e-01 -6.067413e-01 -4.578589e+00
## Cat2.to.1layhid3 -9.629211e-01 1.540598e-01 3.421396e-01
## Intercept.to.1layhid4 -6.348233e-01 -8.025297e-01 8.927208e-01
## Cat1.to.1layhid4 2.498918e-01 7.397678e-02 -4.944209e-02
## Cat2.to.1layhid4 1.946966e-01 1.906542e-01 -3.577438e-01
## Intercept.to.1layhid5 4.784375e+00 -8.633072e-01 -2.621722e+00
## Cat1.to.1layhid5 8.438418e+00 1.892506e-01 4.982159e-01
## Cat2.to.1layhid5 6.644521e+00 -4.987882e-02 6.349624e-01
## Intercept.to.1layhid6 -4.816419e+00 -1.772219e+00 -1.704533e+00
## Cat1.to.1layhid6 -1.267209e-01 3.892254e-01 3.907165e-01
## Cat2.to.1layhid6 6.328688e-01 -1.056768e-01 8.370674e-03
## Intercept.to.1layhid7 7.798317e-01 -1.198510e+00 -8.293560e-02
## Cat1.to.1layhid7 -1.590813e-01 3.111842e-01 9.799909e-02
## Cat2.to.1layhid7 -2.048647e-01 4.985407e-01 -7.211895e-01
## Intercept.to.1layhid8 -1.333428e+00 -3.287892e-01 -4.329648e+00
## Cat1.to.1layhid8 -1.796163e+00 -1.422862e-01 4.232191e-01
## Cat2.to.1layhid8 -2.292374e-01 -1.660481e-01 -4.149045e-02
## Intercept.to.1layhid9 2.875678e+00 3.212453e+01 -4.569664e+00
## Cat1.to.1layhid9 -2.936289e-01 3.275825e-01 -1.544636e+00
## Cat2.to.1layhid9 -1.467669e-01 3.341107e-01 8.238176e-01
## Intercept.to.1layhid10 6.082931e+00 -2.189870e+00 -2.433899e+00
## Cat1.to.1layhid10 -1.195393e+00 -2.883928e-01 3.908976e-01
## Cat2.to.1layhid10 5.471151e-01 5.428211e-01 6.987496e-02
## Intercept.to.Hyp 1.629905e+00 -2.610686e-02 1.958606e+00
## 1layhid1.to.Hyp 4.883070e+00 3.161109e+00 -1.001807e+00
## 1layhid2.to.Hyp -2.505219e+00 5.642248e+00 7.473648e+00
## 1layhid3.to.Hyp -1.135234e+00 -3.267921e+00 -1.055418e+00
## 1layhid4.to.Hyp 6.617429e+00 8.260005e-01 -2.802842e+00
## 1layhid5.to.Hyp 2.073334e+00 1.315910e+00 1.724415e+00
## 1layhid6.to.Hyp 3.921345e+00 6.790403e+00 2.212220e+00
## 1layhid7.to.Hyp -3.244486e+00 1.973456e+00 2.022351e+00
## 1layhid8.to.Hyp 3.469689e+01 -1.978551e+00 5.763455e+00
## 1layhid9.to.Hyp -1.326125e+00 2.170452e+00 1.152008e+00
## 1layhid10.to.Hyp 1.307099e+00 1.491227e+00 3.050540e+00
##
## attr(,"class")
## [1] "nn"
#Estructura de ANN
plot(ann, rep="best")
par(mfrow=c(1,2))
gwplot(ann, selected.covariate = 'Cat1', rep = 'best')
gwplot(ann, selected.covariate = 'Cat2', rep = 'best')
#Predicción
output <- compute(ann, test[ , c("Cat1", "Cat2")], rep = 1)
data.frame(Real = test$Hyp, Predicted = output$net.result, Error = abs(test$Hyp - output$net.result) / test$Hyp)