On Mon, Mar 28, 2011 at 11:01 AM, Adrian May <[email protected]> wrote: > No, it'll be those random numbers again. I'm trying to simulate the road > that my neural network is gonna try to drive along. It should wind back and > forth more or less like real roads. The list I'm making is a set of values > for the curvature at each moment in time (the car's speed is constant for > now.) So each value is a small random change to the previous one, with a bit > of tuning if I can be bothered. I'm thinking of choosing the change in > curvature from a normal distribution biased towards going straight again. I > can't have the curvature go off the scale. Any off-the-shelf normal > distribution functions?
If I were trying to do what I think you are trying to do, I imagine I would use something like: 32 | +/\ (_2+?@#&5) 100 (Here, I have assumed compass points, so 32 and 0 are pointing the same direction. Since this is supposed to represent a road, I must imagine that it has bridges and tunnels such that if the same underlying location appears twice at two different places on the road they are not at the same elevation.) > I just realised that I can't acquire a huge list of snapshots because this > thing might take a hell of a long time to learn. Maybe the printing hack was > right: at least it saved memory. I have no use for the old values. (Personally, I would put off efficiency issues until after I had the code working the way I wanted it to work. I find that manipulating the code is easier when I have a working model to test my thouights.) -- Raul ---------------------------------------------------------------------- For information about J forums see http://www.jsoftware.com/forums.htm
