tobigithub

InChIKey=ACAHMMGNYHEOMN-UHFFFAOYSA-N
InChIKey=AEMHIJYDVNIZCG-UHFFFAOYSA-N
InChIKey=AEMQCMLHPDVYCV-UHFFFAOYSA-N
InChIKey=AEPNDGMPRCFRKR-UHFFFAOYSA-N
InChIKey=AGRCWCGWOIECPK-UHFFFAOYSA-N
InChIKey=AHIPWBCASABPFJ-UHFFFAOYSA-N
InChIKey=AMNJWFPHKGUIDQ-UHFFFAOYSA-N
InChIKey=AMVQEWGAVVIDKE-UHFFFAOYSA-N
InChIKey=AQIDTXWEPULKKH-UHFFFAOYSA-N
InChIKey=ARWIECZMCKTHCZ-UHFFFAOYSA-N

tobigithub

ID#	global	num	formula 	AMG count	formula	AMG isomers	AMG time [sec]	CHECK	CHECK	Formula	No.  Isomers MOLGEN	check formula	check isomer num	C	H	Br	Cl	F	N	O	P	S	Si	Exist	Lewis Check	SeniorMax	DU	# e-	#Atoms	LEWIS Sum	H/C	H/C OK	N/C	O/C	P/C	S/C	NOPS Check	HNOPS Probab	accurate mass	M	M+1	M+2	M+3
1	2	2	C2	0	C2	2	0.33	TRUE	TRUE	C2	0	TRUE	TRUE	2	0	0	0	0	0	0	0	0	0	NO	YES	YES	3	8	2	8	0.00	NO	0.00	0.00	0.00	0.00	YES	NO	24.00000000	100.00	0.00	0.00	0.00
2	3	3	C2H2	1	C2H2	3	0.32	TRUE	TRUE	C2H2	1	TRUE	TRUE	2	2	0	0	0	0	0	0	0	0	YES	YES	YES	2	10	4	10	1.00	YES	0.00	0.00	0.00	0.00	YES	YES	26.01564920	100.00	0.00	0.00	0.00
3	4	4	C2H4	1	C2H4	4	0.32	TRUE	TRUE	C2H4	1	TRUE	TRUE	2	4	0	0	0	0	0	0	0	0	YES	YES	YES	1	12	6	12	2.00	YES	0.00	0.00	0.00	0.00	YES	YES	28.03129840	100.00	0.00	0.00	0.00
4	5	5	CH2O	1	CH2O	5	0.32	TRUE	TRUE	CH2O	1	TRUE	TRUE	1	2	0	0	0	0	1	0	0	0	YES	YES	YES	1	12	4	8	2.00	YES	0.00	1.00	0.00	0.00	YES	YES	30.01056420	100.00	1.15	0.21	0.00
5	6	6	C2H6	1	C2H6	6	0.33	TRUE	TRUE	C2H6	1	TRUE	TRUE	2	6	0	0	0	0	0	0	0	0	YES	YES	Y

tobigithub

#Setup
rm(list = ls(all = TRUE)) 
gc(reset=TRUE)
set.seed(42) #From random.org

#Libraries
library(caret)
library(devtools)
install_github('caretEnsemble', 'zachmayer') #Install zach's caretEnsemble package

tobigithub

Change Point Detection Packages in R
Thanks to the R community, there are packages already existing on CRAN all focusing on change point detection. Especially the following packages are useful because they are not restricted to a special application domain and applicable to time series in general:

CPM – “Parametric and Nonparametric Sequential Change Detection in R”:
Useful for detecting multiple change points in a time series from an unknown underlying distribution. Another bonus is that the method is applicable to data streams, where an observation is only considered once. Because of the “stream nature” of the cpm approach a second output are the detection points themselves. They mark the time when the change point is detected by the algorithm and quantify the delay. Unfortunately the cpm package is no longer maintained on CRAN. For windows users I uploaded a zipped version of the installed package from my R library here. It should work with R 3.0 and 3.1 under Windows 7/8.
BCP – “An R Package for Performi

tobigithub

# things that affect speeds of random forests
# http://stackoverflow.com/questions/14106010/parallel-execution-of-random-forest-in-r/15771458#15771458
# also
# http://stackoverflow.com/questions/23075506/how-to-improve-randomforest-performance?lq=1
# also
#

Setting .multicombine to TRUE can make a significant difference:
````R
rf <- foreach(ntree=rep(25000, 6), .combine=combine, .multicombine=TRUE,

tobigithub

# Run R code on 32 bit and 64 bit and get different results 
# You need to run the code on 32 Bit R version and then the 64 bit R (not just the code below)
# http://stackoverflow.com/questions/17881609/parallel-randomforest-with-different-results-using-dosnow
#
library(foreach)
library(doSNOW)
library(parallel)
set.seed(666)
ncores <- 4
cl <- makeCluster(ncores)

tobigithub

# Random forest are random...indeed
# http://stats.stackexchange.com/questions/35609/why-do-i-need-bag-composition-to-calculate-oob-error-of-combined-random-forest-m
# https://github.com/mlist/IB2014/blob/master/helper_methods.R
# Random Forest combine: http://www.inside-r.org/packages/cran/randomForest/docs/combine
#
# This has implications on parallel random forests using snow, doSNOW, doParallel etc.
#
# err.rate : NULL
# err.rate : NULL
# OOB : NULL

tobigithub

# caret plot functions
# http://machinelearningmastery.com/data-visualization-with-the-caret-r-package/

library(caret)
# load the data
data(iris)

windows()
# pair-wise plots of all 4 attributes, dots colored by class
featurePlot(x=iris[,1:4], y=iris[,5], plot="pairs", auto.key=list(columns=3))

tobigithub

# reproducible caret models
# http://stackoverflow.com/questions/13403427/fully-reproducible-parallel-models-using-caret


library(doParallel); library(caret)

#create a list of seed, here change the seed for each resampling
set.seed(123)
seeds <- vector(mode = "list", length = 11)#length is = (n_repeats*nresampling)+1
for(i in 1:10) seeds[[i]]<- sample.int(n=1000, 3) #(3 is the number of tuning parameter, mtry for rf, here equal to ncol(iris)-2)

tobigithub

# library(memoise) can speed up code 1000x fold
# for true speed-up on fibonacci use library(gmp) fibnum(n)
# http://adv-r.had.co.nz/Function-operators.html
# https://cran.r-project.org/web/packages/memoise/memoise.pdf
# Tobias Kind (2015) https://gist.github.com/tobigithub

library(memoise)

# no memoise()
fib1 <- function(n) {