damonge / cute Goto Github PK

Hello
I have the old version of CUTE working in a local cluster, I moved to the new version for using it in NERSC and the paircounts computed with the new version disagree with old version for the same parameters and xis make no sense, do you have any idea where this could come from?
Mariana

Is there any method to set customize binning of scale instead of set dim1_max and dim1_nbin? For example with monopole correlation function, I want to set spatial scale "r" to estimate the xi(r) at r=np.array([0, 4, 10, 16, 20, 25, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95, 100, 104, 108, 112, 116, 120]) #Mpc/h

Thank you in advance for your reply!

Cheers!

I am getting this error message while trying to run make CUTE

fatal error: gsl/gsl_integration.h: No such file or directory

I have 100s of data files and their corresponding randoms, is there a way to run CUTE on all these with just one command, sort of like a loop command?

Hi, I used CUTE to generate the 3D correlation function in (sigma,pi) pairs. I found that despite its good agreement with the observations, the pair counts in the output column are weighted but not normalised. I'd like to know where I can find the normalisation factors for DD, DR and RR.

Additionally, you had a typo in the LS estimator introduction in CUTE/CUTE/README, which should have been w = (D1D2 - D1R2 - R1D2 + R1R2)/R1R2 but is w = (D1D2 - D1R2 - R1D2 - R1R2)/R1R2 in your code.

Thank you so much for providing such a useful package. Wait for answers.

There is an issue in CUTE/Makefile that the _HAVE_MPI variable is not getting passed when compiling src/io.c. This leads to the MPI_reduce code not getting run and wrong results.

The Makefile runs the following line missing -D_HAVE_MPI

mpicc -Wall -O3 -D_VERBOSE  -D_HAVE_OMP  -D_WITH_WEIGHTS  -D_LOGBIN -DI_R_MAX=.00666 -DLOG_R_MAX=2.176091259 -DI_THETA_MAX=5.72957 -DLOG_TH_MAX=-.758122630 -DI_RL_MAX=.00666 -DI_RT_MAX=.00666 -DI_R3D_MAX=.00666 -DLOG_R3D_MAX=2.176091259 -D_HISTO_2D_64 -DN_LOGINT=10 -c src/io.c -o src/io.o -I./src -I/home/damonge/include

Hi David,

I am trying to run CUTE with my data that is in the format RA DEC z, the first few lines of the file looks like this:

240.691076691 -0.570713581795 0.1943174
244.337480415 -0.429952943145 0.4170079
244.533179409 -0.44337785704 0.4031112
244.770759945 -0.498106983144 0.4358312
245.197502192 -0.479864602544 0.3976909
245.736807492 -0.426961676141 0.4256459
249.521239368 -0.46601278951 0.4227384
237.493807776 -0.0716001579809 0.2745234

When I run cute I got the following error message

Initializing random number generator
First random number : 417175341
Using 1 threads

*** Reading run parameters
Using estimator: LS

*** Setting z-distance relation from the cosmology:

• Omega_M = 0.315
• Omega_L = 0.685
• w = -1.000

*** Monopole correlation function:

• Range: 0.000 < r < 220.000 Mpc/h
• #bins: 50
• Resolution: D(r) = 4.400 Mpc/h
• Using a brute-force approach

*** Reading catalog from file /home/nil/Downloads/DR7radecz_data.txt
30272 lines in the catalog
CUTE: Couldn't read file /home/nil/Downloads/DR7radecz_data.txt, line 1

I have attached the param file.

param.txt

Hello,
I am currently using your code CUTE for my work and I would like to ask you a question about the input parameter 'RR_filename', which makes it possible to feed the random catalogue auto-correlation to the program without the need of recomputing it.
In order to use this feature I added to the parameter_file the following line:

RR_filename = /home/marina/vipers-nn/prova_cf.dat

where /home/marina/vipers-nn/prova_cf.dat is the output of a previous CUTE with the same random catalogue, binning parameters and correlation function type. However, by running the code with the new parameter_file I got the following error:

*** Correlating
 - D-D
Node 0 : 30970 iters, will take from 0 to 30970
    Relative time ellapsed 1497.9 ms
 - D-R
Node 0 : 30970 iters, will take from 0 to 30970
    Relative time ellapsed 3588.9 ms
 - R-R
*** Reading RR from file /home/marina/vipers-nn/prova_cf.dat
CUTE: Couldn't read file /home/marina/vipers-nn/prova_cf.dat, line 1

By changing from w, write, to r, read, at line 192 of the io.c file in the function read_RR, the code runs and reads the RR file:

*** Correlating
 - D-D
Node 0 : 31254 iters, will take from 0 to 31254
    Relative time ellapsed 1476.0 ms
 - D-R
Node 0 : 31254 iters, will take from 0 to 31254
    Relative time ellapsed 3472.5 ms
 - R-R
*** Reading RR from file /home/marina/vipers-nn/prova_cf.dat

    Relative time ellapsed 1.6 ms

*** Writing output file /home/marina/vipers-nn/prova_cf_wRR.dat

*** Cleaning up
             Done !!!             

I just wanted to ask you if I was passing in the wrong way the RR file and so there is no need in making the change in the code, or if the w was just a bug.

Thank you!