HaloFactory - Semi-analytical dark matter halo generators

In astrophysics, it is known that dark matter halos building blocks of the Universe's large-scale structures. Given their significance in theoretical and numerical studies of galaxy formation, it is imperative to generate halo catalogs not only following closely to structure formation theory but also being efficient in computation.

This repository presents a comprehensive collection of semi-analytical methods that facilitate the generation of halo catalogs and their corresponding merger trees. These methods are either purely theoretical or calibrate by numerical simulations. Building upon this repository, we intend to publish another repository dedicated to "semi-analytical galaxy formation models", which will utilize the halo-to-galaxy connection to populate halos with galaxies.

It is important to note that this package is currently in the process of being developed and is subject to structural modifications and functional expansions. The documentation is currently empty, and the testing does not encompass the entire source tree. We will continue to work towards completing these aspects.

Dependencies:

• Python>=3.9.
• pyhipp>=0.0.2, scipy, numpy.

To install halofactory and automatically handle the dependencies, use PyPI:

pip install halofactory

Interesting Features

Halo Density Profile. HaloFactory predefines a set of density profiles for halos. For example, the following codes defines an NFW profile instance:

>>> from halofactory.profile import NFWProfile
>>> profile = NFWProfile(mass = 1.0e2, rs = 10.0)
>>> profile

{'type': 'NFWProfile',
 '_meta': {'mass': 100.0, 'rin': 0.0001, 'xin': 1e-05, 'rvir': 150.0, ... },
 '_us': {'u_l': '3.085677581491367e+19 m', 'u_t': '3.085677581491367e+16 s',
         'u_m': '1.988409870698051e+40 kg', ...}}

The printed message includes the metainfo (mass, virial radius, etc.) of the halo, as well as the unit system in use.

To find information of the profile at given radii, for example, the density, write:

>>> profile.rho([1., 2., 3.])

array([0.03583832, 0.01505707, 0.00855313])

To sample from the profile and thus create a set of particles used as input for N-body simulation, define a sampler such as the Eddington Inversion sampler:

>>> from halofactory.initial_condition import EddInv
>>> sampler = EddInv(profile)
>>> sampler

{'type': 'EddInv',
 '_profile': {'type': 'NFWProfile', ...},
 '_rng': <pyhipp.stats.random.Rng object at 0x15092bc90>,
 '_meta': {'mass': 100.00000000000001, 'Et': 0.0, 'm_hi': 1.0, ...}}

The message indicates the method of the sampler, the input profile, the random number generator, and various of metadata.

To generate a sample, e.g., 1000 particles, write:

>>> sample = sampler.get(1000)
>>> sample.pos, sample.vel, sample.mass

(array([[-27.04020173,  60.70242802,  -6.84922322], ...]),
 array([[  12.7955374 , -114.31584255,   30.89851955],...]),
 0.10000000000000002)

The returned sample instance contains positions, velocities and mass of the sampled particles.

More is coming ...