On the lines where you initialize k and l to [0 0 0 0 0 0], that is an
array of integers by default. Then in the for loops where you try to assign
a floating-point result into that array, you get an InexactError because a
general floating-point number can't be exactly represented as an integer.
If you change those to float64([0 0 0 0 0 0]) or Float64[0 0 0 0 0 0] or
[0.0 0.0 0.0 0.0 0.0 0.0], and likewise with t0, v0, and w0, it should work.
On Sunday, May 4, 2014 2:40:17 PM UTC-7, Paulo Castro wrote:
>
> I was trying to port a function from a previous Python code of mine, and
> ended with this:
>
> # Rubge-Kutta-Fehlberg
> function RKF45(f, g, h, t0, tf, v0, w0)
> # Initial values
> t = t0
> v5 = v4 = v = v0
> w = w0
>
> tolerance = 10^(-5.0)
>
> # Create list for future plots
> t_list = [t]
> v_list = [v]
> w_list = [w]
>
> # Values for a, b and c (as on Butcher's tableau)
> a = [ 0 0 0 0 0 0;
> 1/4 0 0 0 0 0;
> 3/32 9/32 0 0 0 0;
> 1932/2197 -7200/2197 7296/2197 0 0 0;
> 439/216 -8 3680/513 -845/4104 0 0;
> -8/27 2 -3544/2565 1859/4104 -11/40 0]
>
> b = [16/135 0 6656/12825 28561/56430 -9/50 2/55;
> 25/216 0 1408/2565 2197/4104 -1/5 0]
>
> c = [0 1/4 3/8 12/13 1 1/2]
>
> # Relative to y
> k = [0 0 0 0 0 0]
>
> # Relative to z
> l = [0 0 0 0 0 0]
>
> # Compute the next terms
> for i in t0:h:tf
> # Compute the next values of K and L
> for j in 1:6
> k[j] = f(t + c[j] * h, v + (a[j,:] * k')[1], w + (a[j,:] * l'
> )[1])*h
> l[j] = g(t + c[j] * h, v + (a[j,:] * k')[1], w + (a[j,:] * l'
> )[1])*h
> end
>
>
>
> ############################################################################
> # Compute the next value of V
> #
> # Here we implemented a tolerance test
> #
>
> ############################################################################
> v4 = v + (b[2] * k')[1]
> #
> v5 = v + (b[1] * k')[1]
> #
>
> #
> error = abs(v5 - v4)
> #
> if error > tolerance
> #
>
> #
> h = 0.9 * h * ((tolerance/error) ^ (0.25))
> #
>
> #
> for j in 1:6
> #
> k[j] = f(t + c[j] * h, v + (a[j] * k')[1], w + (a[j] * l'
> )[1])*h #
> l[j] = g(t + c[j] * h, v + (a[j] * k')[1], w + (a[j] * l'
> )[1])*h #
> end
> #
>
> #
> v5 = v + (b[1,:] * k')[1]
> #
> end
> #
>
> #
> v = v5
> #
>
> ############################################################################
>
> # Compute T and W with the right values of H and L, obtained after
> the tolerance test
> t += h
> w += (b[1,:] * l')[1]
>
> # Append new values to the lists
> push!(t_list,t)
> push!(v_list,v)
> push!(w_list,w)
> end
>
> return t_list, v_list, w_list
> end
>
> After running this code, I got a mysterious error message:
>
>
>
> *ERROR: InexactError() in setindex! at array.jl:346while loading
> /home/paulo/Documents/Working/ex1.jl, in expression starting on line 28*
>
> ex1.jl is the file I used to test my function:
>
> include("rungekutta.jl")
>
> function f(t, v, w)
> return (v*(v-a)*(1-v) - w + I)/ε
> end
>
> function g(t, v, w)
> return (v - p*w - b)
> end
>
> ε = 0.005
> a = 0.5
> b = 0.15
> p = 1.0
> I = 0.0
> h = 0.005
>
> # Initial and final values of time
> t0 = 0
> tf = 8
>
> t, v, w = RKF45(f, g, h, t0, tf, 0, 0) # v0 = w0 = 0
>
> Can someone help me with finding what's the problem? I tried a lot of
> things, but always end with this error.
>
> Thanks!
>
>
