imeru / mplus Goto Github PK

make one csv file to accumulate result files from each folders.

We need to care about multiple markups such as WALL, WINDOW, EPD, LPD, and etc.

@imeru : Please list MARKUP you need up.

Now, we have only one distribution, one sampling method: Normal distribution-LHS

Combination of distribution and sampling method
distributions: Uniform, Triangular, Normal , Weibull, Gamma, Lognormal and so on.
sampling method: LHS, Random, Quasi-Random

example)
@Wall (n,0,2,20,LHS)
(normal distribution (Gaussian), mean = 0, standard deviation = 2, 20 samples using LHS)
@Wall (u,0,2,20,random)
(uniform distribution, min = 0, max = 2, 20 samples using random)
@Wall (tr,0,2,5,20,quasi-random)
(triangular distribution, min = 0, mode = 2, max = 5, 20 samples using quasi-random)

References------------------------------------------------------------------------------------------------
Random Sampling :
http://docs.python.org/2/library/random.html
http://pymotw.com/2/random/

Latin hyper-cube sampling :
https://mathieu.fenniak.net/latin-hypercube-sampling/
http://pythonhosted.org/pyDOE/randomized.html#latin-hypercube
http://nullege.com/codes/show/src@m@o@model-builder-HEAD@[email protected]
http://stackoverflow.com/questions/12884361/latin-hypercube-sampling-from-a-normal-distribution-python

Quasi Random Sampling :
https://gist.github.com/jkeirstead/1730440 (it's a R code)
http://people.sc.fsu.edu/~jburkardt/py_src/sobol/sobol.html
http://jdherman.github.io/SALib/
http://waterprogramming.wordpress.com/2013/08/05/running-sobol-sensitivity-analysis-using-salib/

• Random Sampling
• Quasi Random Sampling

$ ReadVarsESO my.rvi
After finishing simulation, eplusout.eso should be converted to eplusout.csv using above command. (run ReadVarsESO)

Organize input parameter values in a csv file.
should be separated with result (output) file.

시뮬레이션 돌린후.
인풋 파라미터 벨루를 모아놓은 csv파일을 보니!!!!!!!!!!!!!!!!!!11

각 파라미터에서 랜덤 벨류가 순서대로 나온다. (큰수에서 작은수)
랜덤샘플링의 의미가 없음!

JOB_ID|@@Wall@@|@@WINDOWS@@|@@epd@@
0|0.9099818211155848|1.1838942590757784|4.578937683108727
1|0.8093791483509976|1.1172131759442898|4.359965143132395
2|0.7112319748714843|1.03808686630288|4.512593232484404
3|0.6081054354907885|0.896538274080078|4.609267452677068
4|0.508472636925484|0.8493772806715048|3.965036730556328
5|0.40842714412032394|0.701177930290314|4.491718769877673
6|0.30840602048167565|0.6398242931983869|4.244018597863096
7|0.20779923549962467|0.5131774092760653|4.828089477988538
8|0.10844990054489956|0.4358983469961074|3.5898052125894533
9|0.004578090263261818|0.29177548895905775|3.6982027475908428

어떻게 고쳐야 할까요?