After generating a .class file from the java source, I need to add the following two lines of code to make it work:

.jaddClassPath("/home/shige/mydata/test/Gibbs/out/production/Gibbs")

.jaddClassPath("/home/shige/bin/java/parallelcolt-0.9.4.jar")

skip to main |
skip to sidebar
## Tuesday, June 25, 2013

###
Faster Gibbs sampling MCMC from within R

## Thursday, June 20, 2013

###
Build R with Intel MKL

## Monday, June 10, 2013

###
Using Rcpp in agent-based models

## Sunday, June 09, 2013

###
Ubuntu ...

## Friday, June 07, 2013

###
Some interesting R packages for causal analysis

## Monday, June 03, 2013

###
Symbolic algebra solution for R

## Important Links

## Labels

I find this comparison of MCMC performance between different languages helpful.

After generating a .class file from the java source, I need to add the following two lines of code to make it work:

.jaddClassPath("/home/shige/mydata/test/Gibbs/out/production/Gibbs")

.jaddClassPath("/home/shige/bin/java/parallelcolt-0.9.4.jar")

After generating a .class file from the java source, I need to add the following two lines of code to make it work:

.jaddClassPath("/home/shige/mydata/test/Gibbs/out/production/Gibbs")

.jaddClassPath("/home/shige/bin/java/parallelcolt-0.9.4.jar")

Following the instruction here and here, I built and installed R with the MKL library. Running this benchmark test shows that a significant improvement in performance over the default build (8.782 vs. 33.446 second).

Running the same test on my Windows machine (with an older Core2 chip) took 51 seconds. Following the suggestions and replaced the default "Rblas.dll" with the one provided here, reduced the run-time to 26 seconds.

Running the same test on my Windows machine (with an older Core2 chip) took 51 seconds. Following the suggestions and replaced the default "Rblas.dll" with the one provided here, reduced the run-time to 26 seconds.

Here is an example. The Rcpp code has to be slightly modified to run on my Mint linux machine.

What is the biggest's bug in Ubuntu? Unity! Ubuntu has been down the wrong path too far ... Fortunately we still have Mint and Debian.

A number of new R packages are very helpful for the purpose of conducting causal analysis, including the IUPS package, and the CBPS package.

Ryacas, a interface between R and Yacas, looks like a good solution for symbolic algebra for R.

Subscribe to:
Posts (Atom)

- ADMB
- Advanced R Programming
- Armadillo Library
- AUCTEX
- Bayesian Analysis
- CCPR
- China Migration Project
- CIDR
- Developmental Origins Project
- Doing Bayesian Data Analysis
- Emacs
- ESS
- Estout
- GeoDA
- GLMM wiki
- JAGS
- Journal of Statistical Software
- LaTeX wiki book
- Lazarus
- LibreOffice
- Maxima
- Mplus
- My Home Page
- NetLogo
- OpenMX
- PracTeX
- Queens College
- R
- R bloggers
- R-INLA
- Rcpp
- Research Progress
- RStudio
- ShareLaTeX
- Stan
- Stata
- Stata Journal
- StatET
- TeXstudio
- TUGboat

ADMB
(16)
agent-based model
(4)
air quality
(1)
algebra
(1)
Amazon
(1)
American Community Survey
(1)
aML
(11)
Android
(2)
Apophenia
(3)
Apple
(15)
AUCTeX
(2)
Azure
(1)
Bayesian
(72)
BayesX
(1)
Beamer
(1)
benchmark
(1)
beta regression
(1)
bibliographic software
(10)
bibtex
(3)
big data
(14)
BLAS
(2)
blog
(26)
book
(49)
book reader
(3)
bootstrap
(1)
BUGS
(29)
build dependency
(1)
C
(7)
C++
(22)
causal inference
(9)
ceemple
(1)
China
(6)
Chinese
(18)
Cinnamon
(1)
citation
(1)
Clojure
(5)
cloud computing
(6)
computer
(15)
computing
(2)
console
(1)
CPU
(1)
css
(1)
Cure model
(5)
cygwin
(3)
data entry
(1)
data manipulation
(7)
data science
(10)
database
(2)
Debian
(10)
debug
(4)
demography
(9)
desktop
(2)
diagram
(2)
difference-in-difference
(1)
documentation
(1)
dplyr
(2)
eclipse
(11)
ecology
(1)
econometrics
(2)
Emacs
(43)
ESS
(3)
evolutionary psychology
(2)
exams
(1)
flash drive
(1)
font
(2)
Fortran
(10)
frailty
(3)
fraud
(1)
free books
(2)
Free software
(17)
freepascal
(3)
g++
(5)
GEE
(1)
ggmap
(2)
ggplot2
(15)
GIS
(8)
github
(1)
GLMM
(5)
Gnome
(2)
grant
(1)
graph
(18)
graphics
(4)
gtk
(1)
GUI
(2)
hacker
(1)
Hadoop
(3)
happiness
(1)
high performance
(3)
history
(2)
human nature
(1)
ide
(2)
Incanter
(1)
INLA
(3)
instrumental variables
(2)
interaction effect
(1)
ipython
(3)
IRT
(1)
JAGS
(20)
java
(5)
journal
(10)
Julia
(2)
kindle
(1)
knitr
(4)
kvm
(1)
LaplacesDemon
(2)
Laptop
(16)
Latent Gold
(1)
latent variable
(2)
LaTeX
(81)
legend
(1)
library
(4)
LibreOffice
(3)
linu
(1)
Linux
(97)
Lisp
(7)
literature review
(1)
Lubuntu
(1)
LyX
(1)
machine learning
(1)
make
(1)
map
(12)
marginal effect
(1)
Markdown
(19)
matlab
(3)
maximum likelihood
(2)
MCMC
(10)
methodology
(2)
mind mapping
(2)
Mint Linux
(20)
missing values
(3)
mixture model
(7)
MLwiN
(1)
mono
(1)
monte carlo
(1)
mortality
(1)
Mplus
(7)
multilevel model
(14)
multilevel multiprocess model
(8)
multiple imputation
(4)
natural selection
(1)
netlogo
(3)
network
(1)
news
(1)
NSF
(1)
numerical method
(1)
obesity
(1)
octave
(1)
open access
(1)
open source
(9)
OpenOffice
(9)
output
(3)
PAA
(2)
pandoc
(5)
Paper
(7)
parallel
(2)
pdf
(7)
performance
(14)
phone
(1)
pipeline
(1)
politics
(1)
poststratification
(1)
prediction
(2)
presentation
(3)
prior
(1)
programming
(26)
projection
(1)
propensity score
(1)
proposal
(3)
public use data
(7)
python
(19)
quantile regression
(1)
R
(117)
R packages
(4)
random effect
(3)
Rcpp
(11)
regular expression
(2)
relative distribution
(1)
Repast
(1)
report
(2)
reproducible research
(6)
rescue
(1)
research project
(1)
rpubs
(1)
rstudio
(20)
Russia
(1)
Sabre
(1)
scam
(1)
Sciviews
(2)
scripting
(2)
sequence analysis
(1)
sex ratio
(1)
sex selection
(1)
sharing monitor
(2)
shell
(2)
simulation
(14)
skewed distribution
(1)
skype
(1)
sociology
(4)
software
(20)
solid state drive
(1)
source tarball
(1)
Spark
(7)
SparkR
(8)
spatial analysis
(10)
spell checking
(1)
spline
(1)
ssh
(3)
stan
(15)
Stata
(37)
StatET
(3)
statistics
(4)
structural equation modeling
(5)
summary statistics
(1)
survey data
(1)
survival model
(2)
sweave
(15)
synchronize
(1)
SyncTeX
(1)
table
(23)
tablet pc
(10)
talk
(1)
teaching
(10)
text analysis
(3)
text editor
(8)
text mining
(4)
theme
(1)
tidyr
(1)
tidyverse
(2)
time series
(1)
trust
(1)
ubuntu
(31)
Unix
(2)
Unix philosophy
(1)
version control
(7)
vim
(1)
virtualization
(3)
vmware
(1)
voip
(2)
vpn
(3)
web data
(3)
web log
(1)
web scraping
(2)
wiki
(2)
Windows
(1)
word count
(1)
workflow
(5)
workstation
(2)
writing
(2)
Xfce
(1)
Zelig
(2)
Zotero
(3)