README
Welcome for any question or suggestion: xjc14@mails.tsinghua.edu.cn
Install:
- Pre-rqequist: FFTW3, GSL Library, gcc
- Install: Extract the tar file in WORKDIR, then (in terminal) $ cd Programs $ make test_MMRRM test_powers test_statool test_pdf test_mudecomp test_power_ratio
- If you need re-compile: $ make clean $ make test_MMRRM test_powers test_statool test_pdf test_mudecomp test_power_ratio
Quick Start:
- Edit user changable parameters, flags & definition in WORKDIR/Programs/header/mmrrm_adv.h (You could just turn off all the analysis flags if you only need the output data cube) (NOTE: IF YOU DO NOT HAVE Ts DATA, SET BOTH "OPTHIN" AND "HIGHTS"=1, AND SET Ts DIR IN WORKDIR/prmt/arguments_mmrrm_adv.dat "NULL")
- Re-compile the program
- Edit arguments file WORKDIR/prmt/args_mmrrm_adv.dat
- $ cd Programs/
- $ ./test_MMRRM
- Now the RSD corrected delta_Tb data cube should be in WORKDIR/Output_boxes (Note: the size of delta_Tb data cube is dimdimdim*sizeof(double), so the cube is an double array of shape (dim-x, dim-y, dim-z). )
Source File Configuration:
MMRRM_adv.c: The main program, which mainly operates the MMRRM scheme. There are various of flags in the file "mmrrm_adv.h" in Programs/header, for users to decide whether to turn on some specific function. Some parameters and constants are also defined in "mmrrm_adv.h". The main program would call the following functions (powers, mudecomp, pixelfluxpdf, statool) if needed. That means, you do not need to call these programs deliberately, they are called by test_MMRRM as subroutine.
Usage:
1) Edit the arguments file in ./prmt
See "./prmt/args_mmrrm_adv_instruction.txt" for further explanation.
2) ./test_MMRRM
The program would read-in the data cube according to prmt file, then calculate
RSD corrected data cube and dump them it into ./Output_boxes. After that the main
program would call subroutines to do some statistic analysis, all the analysis
results are dumped into ../Output_PS (by default). For the output format of these
results, see following terms.
powers.c: Calculates the auto power spectrum, in k or log(k) interval. To calculate cross ps, some adaptions are necessary. Relies on FFTW3
Usage(if singly):
./test_powers <Tb along x> <Tb along y> <Tb along z>
<> refers to the file name(WITHOUT PATH, you can edit the path in mmrrm_adv.h) of
brightness temperature output, with mesh to mesh applied along x, y or z direction.
Output format:
linear: kint k_magnitude(1/Mpc) power(mK^2) deviation
log: k_magnitude(1/Mpc) power(mK^2) deviation
mudecomp.c: Calculates the fitting ps for \mu^4 term, where \mu refers to the cosine of the angle between 3-D k vector and LoS. (Yi Mao et.al. 2012) Relies on GSL(SVD fitting) P.S. There's maybe some problems in fitting for log(k) interval, I'll fix it as long as I have time.
Usage(if singly): ./test_mudecomp <Tb along x> <Tb along y> <Tb along z>
Output format:
linear: k(1/Mpc) mu_4(fitting) dev_mu_4(fitting) chi_square P_mu4(quasi-linear) scatter(quasi-linear)
log: k(1/Mpc) mu_4(fitting) dev_mu_4(fitting) chi_square P_mu4(quasi-linear) scatter(quasi-linear)
pixelfluxpdf.c: Calculates the probability distribution function of Tb or other data cube.
Usage(if singly): ./test_pdf <Tb along x> <Tb along y> <Tb along z> [Output file name]
[] means optional
Output format:
delta_Tb[mK] (counts/total_counts)
statool.c: Calculating some statistics: mean, deviation, rms, skewness and kurtosis. in order to read data, this file will read an input list, which records the file name of the input data cube. The main program would generate a list in ./Log_files/Output_List by default.
Usage: ./statool <List name(path also should be included)>
Folders:
Log_files: Record the log file for each program Output_boxes: Record the RSD corrected delta_Tb data cube Output_PS: Record the analysis results Scripts: Record the python scripts used to plot the results as well as the figure prmt: Arguments file for the main program Programs: Source files