CoLoRe is a parallelized code to generate fast 3D realization of a wide variety of cosmological observations.
The methods used by CoLoRe are described in Ramirez-Perez et al. 2021.
When in doubt, bear in mind that by default CoLoRe uses the following units:
- Lenghts: Mpc/h
- Angles: degrees
To compile CoLoRe, open the Makefile and edit it according to your system. The default options (except for the paths to the external libraries) should work for most systems.
The following compilation flags, tunable in the Makefile, can be modified to control the behaviour of CoLoRe:
- CoLoRe may be very memory-demanding. To minimize the memory overhead use single precision floating point (
USE_PRECISION = yes). - OpenMP parallelization is enabled by setting the option
USE_OMPto "yes". - MPI parallelization is enabled by setting the option
USE_MPIto "yes". - The default bias model in CoLoRe is exponential. To select a different bias model add
-D_BIAS_MODEL_2(exp-truncated) or-D_BIAS_MODEL_3(truncated).
CoLoRe uses 4 external packages:
- GSL. The GNU Scientific Library (tested for versions 3.*)
- FFTW. The Fastest Fourier Transform of the West (versions 3.*)
- CFITSIO. FITS format library. This package is optional.
- HDF5. HDF5 format library. This package is optional. The paths to the corresponding headers and libraries should be correctly set in the Makefile.
Once the Makefile has been editted, typing 'make' should generate the executable 'CoLoRe'. To run CoLoRe just type
mpirun -np ./CoLoRe <param_file>
where <param_file> is the path to the parameter file described in section 3.
The behaviour of CoLoRe is mainly controlled by the input param file. The param file is basically a set of name-value pairs. Any blank lines, and anything beyond a #-symbol will be ignored. We provide a sample param file that includes all the input parameters needed by CoLoRe. The comments included in this file explain the meaning and functionality of these parameters.
We also provide an ipython notebook that demonstrates how to generate all the different probes implemented in CoLoRe. The notebook also exemplifies how to interpret the different outputs.
CoLoRe is distributed under the GPL license (see COPYING in the root directory). We kindly ask you to cite the companion paper Ramirez-Perez et al. 2021 when using the code.
Regarding bugs, suggestions, questions or petitions, feel free to contact the authors: David Alonso: [email protected] Cesar Ramirez: [email protected]