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
qrnn trainPredict arguments - inputs x, y
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(qrnn)
qrnn.method <- "none"
hyperParams.qrnn <- function(optim_method, ...) {
iter <- maxit2ndorder
init.range = c(-0.1, 0.1, -0.1, 0.1)
out <- list(iter = iter, init.range=init.range)
return (out)
}
NNtrain.qrnn <- function(x, y, dataxy, formula, neur, optim_method, hyperParams,...) {
hyper_params <- do.call(hyperParams, list(optim_method, ...))
NNreg <- qrnn::qrnn.fit(x, y, n.hidden = neur,
iter.max = hyper_params$iter, n.trials = 1,
init.range = hyper_params$init.range, trace=FALSE)
return (NNreg)
}
NNpredict.qrnn <- function(object, x, ...)
qrnn::qrnn.predict(x, object)
NNclose.qrnn <- function()
if("package:qrnn" %in% search())
detach("package:qrnn", unload=TRUE)
qrnn.prepareZZ <- list(xdmv = "m", ydmv = "m", zdm = "d", scale = TRUE)
if(FALSE)
res <- trainPredict_1data(1, qrnn.method, "NNtrain.qrnn", "hyperParams.qrnn", "NNpredict.qrnn",
NNsummary, "NNclose.qrnn", NA, qrnn.prepareZZ, nrep=5, echo=TRUE, doplot=FALSE,
pkgname="qrnn", pkgfun="qrnn.fit", csvfile=TRUE, rdafile=TRUE, odir=odir)
Launch package’s trainPredict
res <- trainPredict_1pkg(1:12, pkgname = "qrnn", pkgfun = "qrnn.fit", qrnn.method,
prepareZZ.arg = qrnn.prepareZZ, nrep = nrep, doplot = TRUE,
csvfile = TRUE, rdafile = TRUE, odir = odir, echo = FALSE)
Results
#print(res)
kable(t(apply(res, c(1,4), min)))%>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed"))
|
|
RMSE
|
MSE
|
MAE
|
WAE
|
time
|
|
mDette
|
0.3562
|
0.1268
|
0.2420
|
1.7713
|
0.16
|
|
mFriedman
|
0.0121
|
0.0001
|
0.0081
|
0.0413
|
0.28
|
|
mIshigami
|
0.6514
|
0.4244
|
0.3720
|
3.5412
|
0.61
|
|
mRef153
|
3.3111
|
10.9635
|
2.0802
|
13.5466
|
0.15
|
|
uDmod1
|
0.0601
|
0.0036
|
0.0425
|
0.1886
|
0.13
|
|
uDmod2
|
0.0606
|
0.0037
|
0.0428
|
0.1535
|
0.07
|
|
uDreyfus1
|
0.0087
|
0.0001
|
0.0059
|
0.0230
|
0.06
|
|
uDreyfus2
|
0.0967
|
0.0093
|
0.0709
|
0.2487
|
0.08
|
|
uGauss1
|
2.3495
|
5.5204
|
1.8349
|
7.0225
|
0.13
|
|
uGauss2
|
2.4028
|
5.7734
|
1.8803
|
7.9335
|
0.09
|
|
uGauss3
|
2.3342
|
5.4485
|
1.8016
|
7.3949
|
0.13
|
|
uNeuroOne
|
0.2939
|
0.0864
|
0.2258
|
0.7231
|
0.05
|
kable(t(apply(res, c(1,4), median)))%>%
kable_styling(bootstrap_options = c("striped", "hover", "condensed"))
|
|
RMSE
|
MSE
|
MAE
|
WAE
|
time
|
|
mDette
|
1.03775
|
1.07710
|
0.63025
|
7.70380
|
0.390
|
|
mFriedman
|
0.02030
|
0.00040
|
0.01410
|
0.08395
|
0.510
|
|
mIshigami
|
0.79350
|
0.62960
|
0.48760
|
4.04035
|
1.090
|
|
mRef153
|
3.42575
|
11.73640
|
2.16255
|
15.17465
|
0.220
|
|
uDmod1
|
0.14440
|
0.02085
|
0.08440
|
0.51000
|
0.230
|
|
uDmod2
|
0.08465
|
0.00715
|
0.05725
|
0.24415
|
0.240
|
|
uDreyfus1
|
0.30825
|
0.09510
|
0.19510
|
0.94665
|
0.115
|
|
uDreyfus2
|
0.12580
|
0.01580
|
0.09280
|
0.29325
|
0.145
|
|
uGauss1
|
2.73215
|
7.46495
|
2.12470
|
8.70795
|
0.235
|
|
uGauss2
|
7.27225
|
54.16155
|
5.05260
|
21.32695
|
0.240
|
|
uGauss3
|
3.52450
|
12.42345
|
2.54595
|
11.80575
|
0.180
|
|
uNeuroOne
|
0.29390
|
0.08640
|
0.22580
|
0.72310
|
0.075
|