1 Setup

1.1 Packages and options

library(NNbenchmark)
library(kableExtra)
options(scipen = 999)

1.2 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

1.3 validann 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(validann)
validann.method <- c("Nelder-Mead", "BFGS", "CG", "L-BFGS-B", "SANN")
hyperParams.validann <- function(optim_method, ...) {
    
    if (optim_method == "Nelder-Mead")  {iter <- maxit1storderB} 
    if (optim_method == "BFGS")         {iter <- maxit2ndorder}
    if (optim_method == "CG")           {iter <- maxit1storderA}
    if (optim_method == "L-BFGS-B")     {iter <- maxit2ndorder}
    if (optim_method == "SANN")         {iter <- maxit1storderA}
  
    out <- list(iter = iter)
    return (out)
}
NNtrain.validann <- function(x, y, dataxy, formula, neur, optim_method, hyperParams, NNfullformula, NNparam,...) {
    
    hyper_params <- do.call(hyperParams, list(optim_method, ...))
    
    iter <- hyper_params$iter
    method <- hyper_params$method
    
    NNreg <- validann::ann(x, y, size = neur, 
                           method = optim_method, maxit = iter)
    
    return (NNreg)
}
NNpredict.validann <- function(object, x, ...)
  predict(object, x)
NNclose.validann <- function()
  if("package:validann" %in% search())
    detach("package:validann", unload=TRUE)
validann.prepareZZ <- list(xdmv = "m", ydmv = "m", zdm = "d", scale = TRUE)
if(FALSE)
res <- trainPredict_1data(1, validann.method, "NNtrain.validann", "hyperParams.validann", "NNpredict.validann", 
                               NNsummary, "NNclose.validann", NA, validann.prepareZZ, nrep=5, echo=TRUE, doplot=FALSE,
                               pkgname="validann", pkgfun="ann", csvfile=TRUE, rdafile=TRUE, odir=odir)

2 Launch package’s trainPredict

res <- trainPredict_1pkg(1:12, pkgname = "validann", pkgfun = "ann", validann.method,
  prepareZZ.arg = validann.prepareZZ, nrep = nrep, doplot = FALSE,
  csvfile = TRUE, rdafile = TRUE, odir = odir, echo = FALSE)

3 Results

#print(res)
kable(apply(res[,,1,], c(3,1), min), caption = "Nelder-Mead")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

Nelder-Mead
	RMSE	MSE	MAE	WAE	time
mDette	2.6242	6.8863	1.7653	13.4298	1.61
mFriedman	0.0903	0.0082	0.0734	0.2922	12.56
mIshigami	2.4313	5.9112	1.9925	6.6966	28.64
mRef153	3.9160	15.3350	2.8203	12.5384	35.09
uDmod1	0.1364	0.0186	0.0970	0.3993	61.19
uDmod2	0.0813	0.0066	0.0643	0.1852	56.74
uDreyfus1	0.0837	0.0070	0.0507	0.2074	15.02
uDreyfus2	0.1153	0.0133	0.0926	0.2777	16.51
uGauss1	10.2802	105.6825	8.5612	22.6390	46.91
uGauss2	5.0229	25.2296	3.7622	16.6120	42.46
uGauss3	3.7876	14.3458	2.9598	10.3789	33.56
uNeuroOne	0.2983	0.0890	0.2485	0.5724	12.94

kable(apply(res[,,2,], c(3,1), min), caption = "BFGS")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

BFGS
	RMSE	MSE	MAE	WAE	time
mDette	0.2964	0.0878	0.2347	1.2450	1.64
mFriedman	0.0092	0.0001	0.0073	0.0293	2.47
mIshigami	0.6049	0.3659	0.4351	2.8035	5.56
mRef153	3.1802	10.1135	2.1864	12.3262	0.54
uDmod1	0.0412	0.0017	0.0312	0.1132	0.79
uDmod2	0.0427	0.0018	0.0330	0.0869	0.73
uDreyfus1	0.0018	0.0000	0.0013	0.0049	0.23
uDreyfus2	0.0906	0.0082	0.0724	0.2197	0.22
uGauss1	2.2396	5.0156	1.7361	7.0394	1.08
uGauss2	2.3528	5.5357	1.8491	7.4254	0.90
uGauss3	2.3033	5.3050	1.8439	7.0882	0.57
uNeuroOne	0.2830	0.0801	0.2313	0.5675	0.12

kable(apply(res[,,3,], c(3,1), min), caption = "CG")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

CG
	RMSE	MSE	MAE	WAE	time
mDette	0.3785	0.1433	0.2921	1.4510	11.52
mFriedman	0.0176	0.0003	0.0139	0.0555	35.92
mIshigami	0.6883	0.4737	0.5126	2.6717	87.78
mRef153	3.1181	9.7223	2.1554	12.9054	54.27
uDmod1	0.0494	0.0024	0.0393	0.1280	62.23
uDmod2	0.0507	0.0026	0.0418	0.1105	59.14
uDreyfus1	0.0051	0.0000	0.0041	0.0136	49.23
uDreyfus2	0.0909	0.0083	0.0728	0.2231	53.42
uGauss1	2.3153	5.3607	1.8289	7.7608	75.68
uGauss2	2.8878	8.3393	2.3108	8.3963	63.09
uGauss3	2.5744	6.6276	2.0602	7.7276	70.57
uNeuroOne	0.2830	0.0801	0.2313	0.5674	56.35

kable(apply(res[,,4,], c(3,1), min), caption = "L-BFGS-B")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

L-BFGS-B
	RMSE	MSE	MAE	WAE	time
mDette	0.3817	0.1457	0.2978	1.3503	1.76
mFriedman	0.0163	0.0003	0.0131	0.0466	2.75
mIshigami	0.7714	0.5950	0.5692	2.7837	5.70
mRef153	3.1296	9.7943	2.1737	11.1759	1.28
uDmod1	0.0525	0.0028	0.0436	0.1089	1.09
uDmod2	0.0525	0.0028	0.0444	0.1010	0.92
uDreyfus1	0.0015	0.0000	0.0013	0.0037	0.65
uDreyfus2	0.0906	0.0082	0.0722	0.2159	0.24
uGauss1	2.3895	5.7096	1.8696	7.8333	1.16
uGauss2	2.6774	7.1686	2.1646	7.9370	1.01
uGauss3	2.3952	5.7371	1.9224	6.2121	1.14
uNeuroOne	0.2830	0.0801	0.2313	0.5674	0.17

kable(apply(res[,,5,], c(3,1), min), caption = "SANN")%>%
  kable_styling(bootstrap_options = c("striped", "hover", "condensed"))

SANN
	RMSE	MSE	MAE	WAE	time
mDette	3.6109	13.0384	2.6870	17.3460	0.17
mFriedman	0.1395	0.0195	0.1097	0.4286	0.22
mIshigami	3.0705	9.4282	2.5252	9.7793	0.32
mRef153	6.6279	43.9290	4.9132	21.2194	0.22
uDmod1	0.2278	0.0519	0.1804	0.5406	0.20
uDmod2	0.2179	0.0475	0.1730	0.4578	0.17
uDreyfus1	0.1969	0.0388	0.1485	0.5082	0.22
uDreyfus2	0.2425	0.0588	0.1976	0.5504	0.21
uGauss1	12.4076	153.9489	9.3709	32.3143	0.26
uGauss2	10.3689	107.5132	7.8878	25.8157	0.26
uGauss3	5.1887	26.9226	4.1216	15.4642	0.33
uNeuroOne	0.3325	0.1105	0.2726	0.6766	0.32

NNbenchmark Generic Function with validann

All members of GSoC 2019 & 2020

2020/08/07

1 Setup

1.1 Packages and options

1.2 Datasets to test

1.3 validann trainPredict arguments - inputs x, y

2 Launch package’s trainPredict

3 Results