Setup
Packages and options
library(NNbenchmark)
library(kableExtra)
options(scipen = 999)
Datasets to test
NNdataSummary(NNdatasets)
## Dataset n_rows n_inputs n_neurons n_parameters
## 1 mDette 500 3 5 26
## 2 mFriedman 500 5 5 36
## 3 mIshigami 500 3 10 51
## 4 mRef153 153 5 3 22
## 5 uDmod1 51 1 6 19
## 6 uDmod2 51 1 5 16
## 7 uDreyfus1 51 1 3 10
## 8 uDreyfus2 51 1 3 10
## 9 uGauss1 250 1 5 16
## 10 uGauss2 250 1 4 13
## 11 uGauss3 250 1 4 13
## 12 uNeuroOne 51 1 2 7
AMORE trainPredict arguments - inputs
if(dir.exists("D:/GSoC2020/Results/2020run03/"))
{
odir <- "D:/GSoC2020/Results/2020run03/"
}else if(dir.exists("~/Documents/recherche-enseignement/code/R/NNbenchmark-project/NNtempresult/"))
{
odir <- "~/Documents/recherche-enseignement/code/R/NNbenchmark-project/NNtempresult/"
}else
odir <- "~"
nrep <- 10
maxit2ndorder <- 200
maxit1storderA <- 1000
maxit1storderB <- 10000
maxit1storderC <- 100000
#library(AMORE)
hyperParams.AMORE <- function(optim_method, ...) {
if (!is.element(optim_method, c("ADAPTgd", "ADAPTgdwm", "BATCHgd", "BATCHgdwm"))) stop("Invalid Parameters.")
if (optim_method == "ADAPTgd") {iter <- maxit1storderA; lr <- 0.01; momentum <- 0; hidden_activation <- "tansig"}
if (optim_method == "ADAPTgdwm") {iter <- maxit1storderA; lr <- 0.01; momentum <- 0.8; hidden_activation <- "sigmoid"}
if (optim_method == "BATCHgd") {iter <- maxit1storderB; lr <- 0.1; momentum <- 0; hidden_activation <- "tansig"}
if (optim_method == "BATCHgdwm") {iter <- maxit1storderB; lr <- 0.1; momentum <- 0.8; hidden_activation <- "tansig"}
out <- list(iter = iter, lr = lr, momentum = momentum, hidden_activation = hidden_activation)
return (out)
}
NNtrain.AMORE <- function(x, y, dataxy, formula, hidden_neur, optim_method, hyperParams, ...) {
hyper_params <- do.call(hyperParams, list(optim_method, ...))
iter <- hyper_params$iter
lr <- hyper_params$lr
momentum <- hyper_params$momentum
hidden_activation <- hyper_params$hidden_activation
net_structure <- AMORE::newff(n.neurons = c(ncol(x), hidden_neur, 1),
learning.rate.global = lr,
error.criterium = "LMS",
hidden.layer = hidden_activation,
method = optim_method,
momentum.global = momentum)
NNreg <- AMORE::train(net = net_structure,
P = x,
T = y,
error.criterium = "LMS",
report = FALSE,
n.shows = iter,
show.step = 1)
return (NNreg)
}
NNpredict.AMORE <- function(object, x, ...)
AMORE::sim.MLPnet(object$net, x)
NNclose.AMORE <- function()
if("package:AMORE" %in% search())
detach("package:AMORE", unload=TRUE)
AMORE.method <- c("ADAPTgd", "ADAPTgdwm", "BATCHgd", "BATCHgdwm")
AMORE.prepareZZ <- list(xdmv = "d", ydmv = "v", zdm = "d", scale = TRUE)
if(FALSE)
res <- train_and_predict_1data(1, AMORE.method, "NNtrain.AMORE", "hyperParams.AMORE", "NNpredict.AMORE",
NNsummary, "NNclose.AMORE", NA, AMORE.prepareZZ, nrep=2, echo=TRUE,
doplot=FALSE, echoreport=0,
pkgname="AMORE", pkgfun="train", rdafile=TRUE, odir=odir)
Launch package’s trainPredict
res <- trainPredict_1pkg(1:12, pkgname = "AMORE", pkgfun = "train", AMORE.method,
prepareZZ.arg = AMORE.prepareZZ, nrep = nrep, doplot = FALSE,
csvfile = TRUE, rdafile = TRUE, odir = odir, echo = FALSE)
Results
#print(res)
kable(apply(res[,,1,], c(3,1), min), caption = "ADAPTgd")%>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed"))
ADAPTgd
|
|
RMSE
|
MSE
|
MAE
|
WAE
|
time
|
|
mDette
|
1.6532
|
2.7330
|
1.2296
|
7.6420
|
0.11
|
|
mFriedman
|
0.0268
|
0.0007
|
0.0215
|
0.0784
|
0.12
|
|
mIshigami
|
0.6630
|
0.4396
|
0.4887
|
2.5491
|
0.20
|
|
mRef153
|
3.3127
|
10.9738
|
2.3505
|
12.9992
|
0.02
|
|
uDmod1
|
0.2028
|
0.0411
|
0.1602
|
0.4981
|
0.03
|
|
uDmod2
|
0.2537
|
0.0644
|
0.2251
|
0.4973
|
0.02
|
|
uDreyfus1
|
0.3449
|
0.1189
|
0.2661
|
0.7619
|
0.01
|
|
uDreyfus2
|
0.3345
|
0.1119
|
0.2600
|
0.8535
|
0.01
|
|
uGauss1
|
12.1677
|
148.0537
|
9.9965
|
28.2968
|
0.04
|
|
uGauss2
|
7.3477
|
53.9887
|
5.6976
|
21.0875
|
0.04
|
|
uGauss3
|
4.7265
|
22.3395
|
3.7936
|
12.1983
|
0.04
|
|
uNeuroOne
|
0.2954
|
0.0873
|
0.2440
|
0.6419
|
0.01
|
kable(apply(res[,,2,], c(3,1), min), caption = "ADAPTgdwm")%>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed"))
ADAPTgdwm
|
|
RMSE
|
MSE
|
MAE
|
WAE
|
time
|
|
mDette
|
0.3895
|
0.1517
|
0.2986
|
1.6714
|
0.17
|
|
mFriedman
|
0.0439
|
0.0019
|
0.0315
|
0.1660
|
0.19
|
|
mIshigami
|
0.8131
|
0.6612
|
0.5943
|
3.1618
|
0.31
|
|
mRef153
|
3.3399
|
11.1548
|
2.3877
|
12.9869
|
0.04
|
|
uDmod1
|
0.1813
|
0.0329
|
0.1486
|
0.3999
|
0.03
|
|
uDmod2
|
0.1179
|
0.0139
|
0.0997
|
0.2625
|
0.03
|
|
uDreyfus1
|
0.1383
|
0.0191
|
0.1166
|
0.3069
|
0.02
|
|
uDreyfus2
|
0.1856
|
0.0344
|
0.1333
|
0.5125
|
0.01
|
|
uGauss1
|
3.4947
|
12.2133
|
2.5236
|
15.3243
|
0.08
|
|
uGauss2
|
4.8675
|
23.6928
|
3.8736
|
15.6405
|
0.07
|
|
uGauss3
|
4.0064
|
16.0509
|
3.1019
|
11.1709
|
0.07
|
|
uNeuroOne
|
0.2849
|
0.0812
|
0.2282
|
0.6242
|
0.01
|
kable(apply(res[,,3,], c(3,1), min), caption = "BATCHgd")%>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed"))
BATCHgd
|
|
RMSE
|
MSE
|
MAE
|
WAE
|
time
|
|
mDette
|
1.1249
|
1.2654
|
0.8003
|
7.6593
|
1.83
|
|
mFriedman
|
0.0174
|
0.0003
|
0.0136
|
0.0552
|
1.80
|
|
mIshigami
|
2.5010
|
6.2550
|
2.1099
|
6.3023
|
2.58
|
|
mRef153
|
3.3372
|
11.1368
|
2.3747
|
12.8511
|
1.39
|
|
uDmod1
|
0.2777
|
0.0771
|
0.2170
|
0.6637
|
1.76
|
|
uDmod2
|
0.1589
|
0.0253
|
0.1266
|
0.3880
|
1.62
|
|
uDreyfus1
|
0.2360
|
0.0557
|
0.1795
|
0.5969
|
1.36
|
|
uDreyfus2
|
0.2570
|
0.0660
|
0.2028
|
0.6037
|
1.37
|
|
uGauss1
|
12.0712
|
145.7150
|
10.2094
|
25.9779
|
1.68
|
|
uGauss2
|
8.7187
|
76.0160
|
6.4030
|
25.4519
|
1.54
|
|
uGauss3
|
5.1436
|
26.4569
|
4.0217
|
14.4848
|
1.53
|
|
uNeuroOne
|
0.2924
|
0.0855
|
0.2410
|
0.6281
|
1.26
|
kable(apply(res[,,4,], c(3,1), min), caption = "BATCHgdwm")%>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed"))
BATCHgdwm
|
|
RMSE
|
MSE
|
MAE
|
WAE
|
time
|
|
mDette
|
1.8460
|
3.4076
|
1.4793
|
7.6908
|
1.82
|
|
mFriedman
|
0.0174
|
0.0003
|
0.0136
|
0.0546
|
1.85
|
|
mIshigami
|
2.4568
|
6.0360
|
2.0688
|
6.0389
|
2.62
|
|
mRef153
|
3.3346
|
11.1198
|
2.3439
|
13.2163
|
1.42
|
|
uDmod1
|
0.1883
|
0.0354
|
0.1521
|
0.4173
|
1.76
|
|
uDmod2
|
0.2514
|
0.0632
|
0.2243
|
0.4591
|
1.62
|
|
uDreyfus1
|
0.1547
|
0.0239
|
0.1134
|
0.4293
|
1.36
|
|
uDreyfus2
|
0.1825
|
0.0333
|
0.1350
|
0.5424
|
1.36
|
|
uGauss1
|
11.9912
|
143.7900
|
10.1183
|
25.8322
|
1.68
|
|
uGauss2
|
8.4071
|
70.6802
|
6.2189
|
24.4918
|
1.55
|
|
uGauss3
|
5.1542
|
26.5656
|
4.0282
|
14.5604
|
1.56
|
|
uNeuroOne
|
0.2927
|
0.0857
|
0.2413
|
0.6289
|
1.27
|