/gravhopper

Simple N-body code for Python (backend is in C for speed) with BH tree and direct summation options.

Primary LanguagePythonBSD 3-Clause "New" or "Revised" LicenseBSD-3-Clause

GravHopper

"They told me computers could only do arithmetic." -- Grace Hopper

Gravitational N-body simulation code written by Jeremy Bailin.

Named in honour of pioneering computer scientist Grace Hopper. Doubly appropriate because it uses a leapfrog integrator.

This is a simple Python-interface code for performing gravitational N-body simulations. It combines a simple Python interface for ease of use with a C backend for speed, and has the following features:

  • Choice of Barnes-Hut tree or direct summation algorithm.
  • Ability to include external potentials from galpy, gala, or user-supplied functions.
  • Ability to return output as pynbody snapshots.
  • Functions that generate equilibrium or near-equilibrium initial conditions (ICs) for several density profiles (Plummer, Hernquist, exponential disk), along with the ability to create ICs from galpy distribution function objects or pynbody snapshots.
  • Utility functions for plotting snapshots and making movies.

For now, it uses a constant uniform timestep and constant uniform Plummer softening.

Requirements:

  • Astropy
  • NumPy, SciPy, Matplotlib
  • C compiler
  • To use galpy, gala, or pynbody interface functions, they will need to be installed.
  • Saving movies requires ffmpeg.

For example, this will create a Plummer sphere with 2000 particles, run it for a few dynamical times, and plot the particle positions before and after to show that it is in equilibrium:

from gravhopper import Simulation, IC
from astropy import units as u
import matplotlib.pyplot as plt

# Create Plummer initial conditions.
Plummer_IC = IC.Plummer(N=2000, b=1*u.pc, totmass=1e6*u.Msun)

# Create a new simulation with a time step of 0.005 Myr and a softening of 0.05 pc.
sim = Simulation(dt=0.005*u.Myr, eps=0.05*u.pc)
# Add the Plummer model to the simulation
sim.add_IC(Plummer_IC)
# Run for 400 time steps
sim.run(400)

# Plot the x-y positions at the beginning and end.
fig = plt.figure(figsize=(12,4))
ax1 = fig.add_subplot(121, aspect=1.0)
ax2 = fig.add_subplot(122, aspect=1.0)
sim.plot_particles(snap='IC', unit=u.pc, xlim=[-10,10], ylim=[-10,10], ax=ax1)
sim.plot_particles(snap='final', unit=u.pc, xlim=[-10,10], ylim=[-10,10], ax=ax2)

To make a movie of the whole evolution of the simulation:

# Make and save a movie of the simulation running
sim.movie_particles('Plummer_sim.mp4', unit=u.pc, xlim=[-10,10], ylim=[-10,10])

Documentation

Full documentation, including installation instructions, examples, and library reference, are at Read The Docs

Installation

Option 1: Pip

If all goes well, you should be able to install GravHopper simply with:

pip install gravhopper

Option 2: Install from source

To install directly from the current source:

  1. Clone or download the git repository
git clone https://github.com/jbailinua/gravhopper.git
  1. Go into the gravhopper directory and build the code.
cd gravhopper
python setup.py build_ext --inplace
  1. Copy the gravhopper subdirectory to wherever you want to use it.