[SciPy-user] Problems with scipy.integrate.ode

[David Linke]

Hi!

I am having problems with the set_f_params() method in ode.py.

An example showing the use of set_f_params() to supply additional 
arguments (two or more) to f would be very heplful.

I tried a while to get it going but without success. I observed strange 
behaviour when using a modified function test1 in ode.py to receive 
additional arguments:

def test1():
     #def f(t,y):
     def f(t,y,b):
         print type(b), b
         a = Numeric.sin(6*t)
         return y*y-a+y
     ode_runner = ode(f)
     ode_runner.set_integrator('vode')
     ode_runner.set_initial_value([0.1,0.11,.1]*10)
     par_b = 0.1,
     ode_runner.set_f_params(par_b)
     while ode_runner.successful() and ode_runner.t < 20:
         y1 = ode_runner.integrate(ode_runner.t+2)
         print ode_runner.t,y1[:3]

Is this the right way to use set_f_params?  When I run this example the 
arguments are passed to f only in the first two calls, as shown here:

Python 2.2.2 (#37, Oct 14 2002, 17:02:34) [MSC 32 bit (Intel)] on win32
 >>> Integrators available: vode
Found integrator vode
<type 'tuple'> (0.10000000000000001,)        ! 1st call
...
2.0 [-0.1446766  -0.10226142 -0.1446766 ]
<type 'tuple'> (0.10000000000000001,)        ! 2nd call
...
4.0 [-0.6351882 -0.5874518 -0.6351882]
<type 'tuple'> (<NULL>,)                     ! 3rd call
...
6.0 [-0.93274299 -0.92029838 -0.93274299]
<type 'tuple'> (<NULL>,)
8.0 [-1.06178405 -1.05979417 -1.06178405]
...


Thank you for any help.

David