Odespy

"Odespy (ODE Software in Python) offers a unified interface to a large collection of software for solving systems of ordinary differential equations (ODEs). There is also some support for Differential Algebraic Equations (DAEs).

The features include:

• Pure Python implementations of classical explicit schemes such as the Forward Euler method (also called Euler); Runge-Kutta methods of 2nd, 3rd, and 4th order; Heun's method; Adams-Bashforth methods of 2nd, 3rd, and 4th order; Adams-Bashforth-Moulton methods of 2nd and 3rd order.
• Pure Python implementations of classical implicit schemes such as Backward Euler; 2-step backward scheme; the theta rule; the Midpoint (or Trapezoidal) method.
• Pure Python implementations of adaptive explicit Runge-Kutta methods of type Runge-Kutta-Fehlberg of order (4,5), Dormand-Prince of order (4,5), Cash-Karp of order (4,5), Bogacki-Shampine of order (2,3).
• Wrappers for all FORTRAN solvers in ODEPACK.
• Wrappers for the wrappers of FORTRAN solvers in scipy: vode and zvode (adaptive Adams or BDF from vode.f); dopri5 (adaptive Dormand-Prince method of order (4,5)); dop853 (adaptive Dormand-Prince method of order 8(5,3)); odeint (adaptive Adams or BDF, basically the same as vode, but in the implementation lsoda from ODEPACK).
• Wrapper for the Runge-Kutta-Chebyshev formulas of order 2 as offered by the well-known FORTRAN code rkc.f.
• Wrapper for the Runge-Kutta-Fehlberg method of order (4,5) as provided by the well-known FORTRAN code rkf45.f.
• Wrapper for the Radau5 method as provided by the well-known FORTRAN code radau5.f. There have been some unidentified problems with running this solver (segmentation fault).
• Wrapper for some solvers in the odelab.

The ODE problem can always be specified in Python, but for wrappers of FORTRAN codes one can also implement the problem in FORTRAN and avoid callback to Python."

https://github.com/hplgit/odespy

http://hplgit.github.io/odespy/doc/web/index.html