bfonta/darkmatter

Dark matter simulation studies performed at Swinburne University of Technology, Melbourne

dmprofile

Current release: v1.0.0

This Python package provides intuitive methods that perform some of the most common steps needed to study dark matter simulations using the pynbody package.

Installing

To use dmprofile you will need the pynbody package. Please follow these instructions. To install dmprofile you have two options:

• Clone the repository:

git clone https://github.com/b-fontana/DarkMatter.git
cd DarkMatter/

and try the examples, which are located in DarkMatter/dmprofile/examples. A data sample composed of subhalos and corresponding snapshot is provided in the data/ folder, so that one can run the examples and scripts and understand the basic workflow.

• Pip install (installs the latest release, which does not necessarily correspond to the code in this repository):

pip install dmprofile

Classes

• Halos(): performs the management of the halos stored in the simulations. The class was built such that pynbody never has to be explicitly invoked. The following code sample:

h = Halos('halos_path', 'snapshot_path', min_size=300)  

will load all the halos with at least 300 particles and corresponding subhalos stored in halos_path and the snapshots (optional) stored in snapshot_path. One can use the argument N instead if a fixed number of halos is preferred. Mixing both N and min_size is also possible.

The instantiation will create backup halos and subhalos in case the user later decides to filter them, making the recover of the original halos possible.

All operations can be performed directly using the instance of the Halos() class. For instance:

h.concentration_200(idx=idx, sub_idx=sub_idx)
h.get_shape(idx=idx, sub_idx=sub_idx)

where idx and sub_idx are the indexes stored internally by pynbody. They are ordered according to the size of the halos. Inside a specific halo, the subhalo having sub_idx=0 will be the largest subhalo, the subhalo with sub_idx=1 will be the second largest, and so on.

For a more comprehensive example, please see script1.sh, or check the examples in the /dmprofile/examples/ folder. To run them, just do

python dmprofile/examples/<name of the example>` 

from a directory placed above the package directory.

• Plot(): Takes care of all the plotting steps. Please see script1_plot.py.

This class is currently able to plot:

• Shape parameters
• Concentration
• Profiles (density, mass, ...)
• Relaxation
• Simple particle distributions

Example of a plot produced with script1_plot.py, displaying shape parameters of several simulations:

• Other utilities:

1. Parser that can be extended, stored in dmprofile/src/parser. To use it and print the parsed arguments:

FLAGS, _ = add_args(argparse.ArgumentParser())
print("Parsed arguments:")
for k,v in FLAGS.__dict__.items():
    print('{}: {}'.format(k,v))

2. General utilities like memory consumption check, I/O operations, and others. Stored in dmprofile/src/utilities.py.

3. NFW_fit (dmprofile/src/fit.py) and centering function for the whole simulation and individual halos and subhalos (dmprofile/src/move.py). The latter file includes an additional function that is currently not operational.

Performing customized operations

In case the user wants to perform some operation that is still not included in the dmprofile package, it can retrieve the wanted halo or subhalo in the following way:

h.get_halo(idx=idx)
h.get_subhalo(idx=idx, sub_idx=sub_idx)

h.get_halos()
h.get_subhalos()

where the last two lines retrieve all the available halos and subhalos, respectively. One can then directly apply pynbody methods to the halos.

Please consider contributing to this package with your own code. I will be glad to review any pull request.

Acknowledgements

This package was developed during an internship at Swinburne University of Technology, Melbourne. I thank my supervisor, Dr. Alan R. Duffy, for having given all the required technical and theoretical support.

Contact: [email protected]