Python binding for odeint from boost.
The Rosenbrock4 stepper requires that the user provides a routine for calculating the Jacobian.
Autogenerated API documentation for latest stable release is found here: https://bjodah.github.io/pyodeint/latest (and the development version for the current master branch are found here: http://hera.physchem.kth.se/~pyodeint/branches/master/html).
Simplest way to install is to use the conda package manager:
$ conda install -c bjodah pyodeint pytest $ python -m pytest --pyargs pyodeint
tests should pass.
Binary distribution is available here: https://anaconda.org/bjodah/pyodeint
Source distribution is available here: https://pypi.python.org/pypi/pyodeint
here is an example of how to build from source:
$ CPLUS_INCLUDE_PATH=~/Downloads/boost_1_63_0 python3 setup.py build_ext -i
The classic van der Pol oscillator (see examples/van_der_pol.py)
>>> from pyodeint import integrate_adaptive # also: integrate_predefined >>> mu = 1.0 >>> def f(t, y, dydt): ... dydt = y ... dydt = -y + mu*y*(1 - y**2) ... >>> def j(t, y, Jmat, dfdt, fy=None): ... Jmat[0, 0] = 0 ... Jmat[0, 1] = 1 ... Jmat[1, 0] = -1 -mu*2*y*y ... Jmat[1, 1] = mu*(1 - y**2) ... dfdt = 0 ... dfdt = 0 ... >>> y0 = [1, 0]; tend=10.0; dt0=1e-8; t0=0.0; atol=1e-8; rtol=1e-8 >>> tout, yout, info = integrate_adaptive(f, j, y0, t0, tend, dt0, atol, rtol, ... method='rosenbrock4', nsteps=1000) >>> import matplotlib.pyplot as plt >>> series = plt.plot(tout, yout) >>> plt.show() # doctest: +SKIP
For more examples see examples/, and rendered jupyter notebooks here: http://hera.physchem.kth.se/~pyodeint/branches/master/examples
pyodesys for how to automatically generate the jacobian callback function (and easily swtich to other solvers).
The source code is Open Source and is released under the very permissive “simplified (2-clause) BSD license”. See LICENSE for further details. Contributors are welcome to suggest improvements at https://github.com/bjodah/pyodeint