Reset system and rerun with different whitenoise signal

[studywolf]

I was wanting to run a single simulation a bunch of times, with the same neurons but with random noise that was different between runs. sim.reset() also reset the noise, so I used this workaround:

other_t = np.random.random()

with m:

    ...

    from nengo.utils.functions import whitenoise
    noise_func = whitenoise(0.1, 10, .001, dimensions=2)
    def func(t):
        print other_t
        return noise_func(t + other_t)
    noisenode = nengo.Node(output=func)

    ...

sim = nengo.Simulator(m)
for ii in range(50):
    sim.reset()
    other_t = np.random.random()
    sim.run(.75)

So this is mostly just a workaround post in case someone is looking to do this. It would be cool if there was a less hack way to do this as well!

[hunse]

The issue is that whitenoise generates the input signal when it's called, not at runtime. A straightforward way to do what you want is to change the node function to a whitenoise function with a different seed, reset the simulator, and off you go. #372 allows changing node functions like this.