Plotting iteratively defined function in MATLAB - matlab

I'm not a MATLAB professional and so, I need some help at producing a 3D plot of the iteratively defined function f : R^2-0 -> R defined below (in pseudocode) for x,y values in [-1,1] and
For each I = 1,2,3 and A = (0.5,0.5)
function f(v in R^2-0)
{
a=b=0; (a,b in R)
for (i=0; i<I; i=i+1)
{
v = |v| / ||v||^2 - A;
a = a + | ||v||-b |;
b = ||v||;
}
return a;
}
(|v| denotes component-wise vector absolute value)
(If you want you can look at the fractal that is generated by the function at my question on math-exchange here:
https://math.stackexchange.com/questions/1457733/a-question-about-a-fractal-like-iteratively-defined-function
)
MATLAB code to do that will be appreciated.
Much Thanks.

Save this your main program:
clear
clc
close all
% I = 1;
% A = [ 0.5 0.5 ];
I = 10;
A = [ 0.5 0.5 0.5 ];
xmin = -1;
xmax = 1;
ymin = -1;
ymax = 1;
nx = 101;
ny = 101;
dx = (xmax - xmin) / (nx - 1);
dy = (ymax - ymin) / (ny - 1);
x = xmin: dx: xmax;
y = ymin: dy: ymax;
for ix = 1: nx
for iy = 1: ny
if (length(A) == 2)
z(iy, ix) = f([x(ix) y(iy)], A, I);
elseif (length(A) == 3)
z(iy, ix) = f([x(ix) y(iy) 0], A, I);
end
end
end
pcolor(x, y, z)
shading interp
Then save this function in the same directory as the main program as f.m:
function result = f(v, A, I)
a = 0;
b = 0;
for i = 1: I
v = abs(v) / dot(v, v) - A;
a = a + abs(norm(v) - b);
b = norm(v);
end
result = a;
end

Related

Steepest Descent using Armijo rule

I want to determine the Steepest descent of the Rosenbruck function using Armijo steplength where x = [-1.2, 1]' (the initial column vector).
The problem is, that the code has been running for a long time. I think there will be an infinite loop created here. But I could not understand where the problem was.
Could anyone help me?
n=input('enter the number of variables n ');
% Armijo stepsize rule parameters
x = [-1.2 1]';
s = 10;
m = 0;
sigma = .1;
beta = .5;
obj=func(x);
g=grad(x);
k_max = 10^5;
k=0; % k = # iterations
nf=1; % nf = # function eval.
x_new = zeros([],1) ; % empty vector which can be filled if length is not known ;
[X,Y]=meshgrid(-2:0.5:2);
fx = 100*(X.^2 - Y).^2 + (X-1).^2;
contour(X, Y, fx, 20)
while (norm(g)>10^(-3)) && (k<k_max)
d = -g./abs(g); % steepest descent direction
s = 1;
newobj = func(x + beta.^m*s*d);
m = m+1;
if obj > newobj - (sigma*beta.^m*s*g'*d)
t = beta^m *s;
x = x + t*d;
m_new = m;
newobj = func(x + t*d);
nf = nf+1;
else
m = m+1;
end
obj=newobj;
g=grad(x);
k = k + 1;
x_new = [x_new, x];
end
% Output x and k
x_new, k, nf
fprintf('Optimal Solution x = [%f, %f]\n', x(1), x(2))
plot(x_new)
function y = func(x)
y = 100*(x(1)^2 - x(2))^2 + (x(1)-1)^2;
end
function y = grad(x)
y(1) = 100*(2*(x(1)^2-x(2))*2*x(1)) + 2*(x(1)-1);
end

Plotting level curves in MATLAB for a complex function

I am trying to plot the contour plots for the given function
syms r x y k z
[ph,r] = meshgrid((0:5:360)*pi/180,0:.5:10);
[X,Y] = pol2cart(ph,r);
Z = X+i*Y;
J = besselj(k,l.*r);
J2 = besselj(k,m.*r);
Y = bessely(k,l.*r);
Y2 = bessely(k,m.*r);
H = besselh(k,r);
F1 = symsum((J).*exp(1i*k*ph),k,-5,5);
F2 = symsum((J2+Y2).*exp(1i.*k.*ph),k,-5,5);
F3 = symsum(H.*exp(1i.*k.*ph),k,-5,5);
pwu = nan(size(F1), 'like', F1);
mask = 0 <= r & r < 0.5;
pwu(mask) = F1(mask);
mask = 0.5 <= r & r < 1;
pwu(mask) = F2(mask);
mask = r >= 1;
pwu(mask) = F3(mask);
U = subs(pwu, {l, m}, {1.5, 3});hold on
contour(X,Y,imag(double(U)),30)
axis equal
xlabel('r','FontSize',14);
ylabel('phi','FontSize',14);
but I keep getting errors for the form of the 4th last line. Apparently U cannot be converted to a double. Is there any other way to do this?
Thanks
This code works:
syms r x y k z
[ph,r] = meshgrid((0:5:360)*pi/180,0:.5:10);
[x,y] = pol2cart(ph,r);
Z = x+1i*y;
J = besselj(k,l.*r);
J2 = besselj(k,m.*r);
Y = bessely(k,l.*r);
Y2 = bessely(k,m.*r);
H = besselh(k,r);
F1 = symsum((J).*exp(1i*k*ph),k,-5,5);
F2 = symsum((J2+Y2).*exp(1i.*k.*ph),k,-5,5);
F3 = symsum(H.*exp(1i.*k.*ph),k,-5,5);
pwu = nan(size(F1), 'like', F1);
mask = 0 <= r & r < 0.5;
pwu(mask) = F1(mask);
mask = 0.5 <= r & r < 1;
pwu(mask) = F2(mask);
mask = r >= 1;
pwu(mask) = F3(mask);
U = subs(pwu, {l, m}, {1.5, 3});hold on
J0 = besselj(k,r);
u0 = symsum(1i.^(-k).*J0.*exp(1i*k*ph),k,-5,5);
W = U+u0;
contour(x, y, angle(double(W)), 30);
axis equal
xlabel('r','FontSize',14);
ylabel('phi','FontSize',14);
Thanks for your comments!

Using a simple Euler method to solve the EoS of Stars (Mathematica)

I attempted to translate a matlab code:
clear
clc;
%Use a resolution of dr = 10^-N
N = 4;
%G = GAMMA, adiabatic parameter
G = 4/3; % high density/extreme relativistic
%G = 5/3; % low density/nonrelativistic
r = 0;
dr = 10^-N;
m(1) = 0;
r(1) = 0;
rho(1) = 0.1;
i = 1;
while (i>0)
i=i+1;
r(i) = r(i-1)+dr;
Gamma(i) = rho(i-1)^(G-1) / (3*sqrt(2)) ;%(rho(i-1))^(2/3) / (3*(1+rho(i-1)^(2/3) )^(1/2) );
drho = (m(i-1)*rho(i-1)) / (Gamma(i) * r(i)^2)*dr;
if (drho > rho(i-1))
break;
else
rho(i) = rho(i-1) - drho;
end
m(i) = m(i-1) + (r(i)^2 * rho(i)) * dr;
end
Mass = m(i-1)
Rad = r(i-1)
plot(r(1:i-1),m(1:i-1))
xlabel('Log10(r/[m])')
ylabel('M(r) (M_{sol})')
Into Mathematica, what I have is:
EoS[n0_] :=
Module[{n = n0},
N1 = 4;
G1 = 4/3;
G2 = 5/3;
rho[n_] := 0.1 + n;
m1[n_] := n;
dr = 10^(-N1);
r[n_] := n;
Gamma1[n_] := ((rho[n - 1])^(G1 - 1))/(3*Sqrt[2]);
drho = ((m1[n - 1]*rho[n - 1]))/(Gamma1[n - 1]*r[n - 1]^2)*dr;
n = 1;
While[n > 0, Gamma1[n] && drho; n++]
If[drho > rho[-1], Break[], rho[n] = rho[-1] - drho]
m1[n] = m1[-1] + ((r[n]^2)*rho[n])*dr;
Mass = m1[n - 1];
Rad = r[n - 1];
ParametricPlot[{m[n], r[n]}, {n, 1, n - 1},
AxesLabel -> {"Log10(r/[m])", "M(r)(M_{sol})"}]
]
My problem is I'm getting complex solutions when I obviously shouldn't. I'm an absolute novice to mathematica, especially with modules and the euler method itself. Any and all feedback/help would be much appreciated. I'm assuming my increments are wrong, I don't really know.

Undefined function or variable 'x'

I am trying to evaluate two matrixes which I defined outside of the function MetNewtonSist using subs and I get the error Undefined function or variable 'x' whenever I try to run the code.
[edit] I added the code for the GaussPivTot function which determines the solution of a liniear system.
syms x y
f1 = x^2 + y^2 -4;
f2 = (x^2)/8 - y;
J = jacobian( [ f1, f2 ], [x, y]);
F = [f1; f2];
subs(J, {x,y}, {1, 1})
eps = 10^(-6);
[ x_aprox,y_aprox, N ] = MetNewtonSist( F, J, 1, 1, eps )
function [x_aprox, y_aprox, N] = MetNewtonSist(F, J, x0, y0, eps)
k = 1;
x_v(1) = x0;
y_v(1) = y0;
while true
k = k + 1;
z = GaussPivTot(subs(J, {x, y}, {x_v(k-1), y_v(k-1)}),-subs(F,{x, y}, {x_v(k-1), y_v(k-1)}));
x_v(k) = z(1) + x_v(k-1);
y_v(k) = z(1) + y_v(k-1);
if norm(z)/norm([x_v(k-1), y_v(k-1)]) < eps
return
end
end
N = k;
x_aprox = x_v(k);
y_aprox = y_v(k);
end
function [x] = GaussPivTot(A,b)
n = length(b);
A = [A,b];
index = 1:n;
for k = 1:n-1
max = 0;
for i = k:n
for j = k:n
if A(i,j) > max
max = A(i,j);
p = i;
m = j;
end
end
end
if A(p,m) == 0
disp('Sist. incomp. sau comp. nedet.')
return;
end
if p ~= k
aux_line = A(p,:);
A(p,:) = A(k, :);
A(k,:) = aux_line;
end
if m ~= k
aux_col = A(:,m);
A(:,m) = A(:,k);
A(:,k) = aux_col;
aux_index = index(m);
index(m) = index(k);
index(k) = aux_index;
end
for l = k+1:n
M(l,k) = A(l,k)/A(k,k);
aux_line = A(l,:);
A(l,:) = aux_line - M(l,k)*A(k,:);
end
end
if A(n,n) == 0
disp('Sist. incomp. sau comp. nedet.')
return;
end
y = SubsDesc(A, A(:,n+1));
for i = 1:n
x(index(i)) = y(i);
end
end
By default, eps is defined as 2.2204e-16 in MATLAB. So do not overwrite it with your variable and name it any word else.
epsilon = 1e-6;
Coming to your actual issue, pass x and y as input arguments to the MetNewtonSist function. i.e. define MetNewtonSist as:
function [x_aprox, y_aprox, N] = MetNewtonSist(F, J, x0, y0, epsilon, x, y)
%added x and y and renamed eps to epsilon
and then call it with:
[x_aprox, y_aprox, N] = MetNewtonSist(F, J, 1, 1, epsilon, x, y);

Vectorizing 4 nested for loops

I'm trying to vectorize the 2 inner nested for loops, but I can't come up with a way to do this. The FS1 and FS2 functions have been written to accept argument for N_theta and N_e, which is what the loops are iterating over
%% generate regions
for raw_r=1:visual_field_width
for raw_c=1:visual_field_width
r = raw_r - center_r;
c = raw_c - center_c;
% convert (r,c) to polar: (eccentricity, angle)
e = sqrt(r^2+c^2)*deg_per_pixel;
a = mod(atan2(r,c),2*pi);
for nt=1:N_theta
for ne=1:N_e
regions(raw_r, raw_c, nt, ne) = ...
FS_1(nt-1,a,N_theta) * ...
FS_2(ne-1,e,N_e,e0_in_deg, e_max);
end
end
end
end
Ideally, I could replace the two inner nested for loops by:
regions(raw_r,raw_c,:,:) = FS_1(:,a,N_theta) * FS_2(:,N_e,e0_in_deg,e_max);
But this isn't possible. Maybe I'm missing an easy fix or vectorization technique? e0_in_deg and e_max are parameters.
The FS_1 function is
function h = FS_1(n,theta,N,t)
if nargin==2
N = 9;
t=1/2;
elseif nargin==3
t=1/2;
end
w = (2*pi)/N;
theta = theta + w/4;
if n==0 && theta>(3/2)*pi
theta = theta - 2*pi;
end
h = FS_f((theta - (w*n + 0.5*w*(1-t)))/w);
the FS_2 function is
function g = FS_gne(n,e,N,e0, e_max)
if nargin==2
N = 10;
e0 = .5;
elseif nargin==3
e0 = .5;
end
w = (log(e_max) - log(e0))/N;
g = FS_f((log(e)-log(e0)-w*(n+1))/w);
and the FS_f function is
function f = FS_f(x, t)
if nargin<2
t = 0.5;
end
f = zeros(size(x));
% case 1
idx = x>-(1+t)/2 & x<=(t-1)/2;
f(idx) = (cos(0.5*pi*((x(idx)-(t-1)/2)/t))).^2;
% case 2
idx = x>(t-1)/2 & x<=(1-t)/2;
f(idx) = 1;
% case 3
idx = x>(1-t)/2 & x<=(1+t)/2;
f(idx) = -(cos(0.5*pi*((x(idx)-(1+t)/2)/t))).^2+1;
I had to assume values for the constants, and then used ndgrid to find the possible configurations and sub2ind to get the indices. Doing this I removed all loops. Let me know if this produced the correct values.
function RunningFunction
%% generate regions
visual_field_width = 10;
center_r = 2;
center_c = 3;
deg_per_pixel = 17;
N_theta = 2;
N_e = 5;
e0_in_deg = 35;
e_max = 17;
[raw_r, raw_c, nt, ne] = ndgrid(1:visual_field_width, 1:visual_field_width, 1:N_theta, 1:N_e);
ind = sub2ind(size(raw_r), raw_r, raw_c, nt, ne);
r = raw_r - center_r;
c = raw_c - center_c;
% convert (r,c) to polar: (eccentricity, angle)
e = sqrt(r.^2+c.^2)*deg_per_pixel;
a = mod(atan2(r,c),2*pi);
regions(ind) = ...
FS_1(nt-1,a,N_theta) .* ...
FS_2(ne-1,e,N_e,e0_in_deg, e_max);
regions = reshape(regions, size(raw_r));
end
function h = FS_1(n,theta,N,t)
if nargin==2
N = 9;
t=1/2;
elseif nargin==3
t=1/2;
end
w = (2*pi)./N;
theta = theta + w/4;
theta(n==0 & theta>(3/2)*pi) = theta(n==0 & theta>(3/2)*pi) - 2*pi;
h = FS_f((theta - (w*n + 0.5*w*(1-t)))/w);
end
function g = FS_2(n,e,N,e0, e_max)
if nargin==2
N = 10;
e0 = .5;
elseif nargin==3
e0 = .5;
end
w = (log(e_max) - log(e0))/N;
g = FS_f((log(e)-log(e0)-w*(n+1))/w);
end
function f = FS_f(x, t)
if nargin<2
t = 0.5;
end
f = zeros(size(x));
% case 1
idx = x>-(1+t)/2 & x<=(t-1)/2;
f(idx) = (cos(0.5*pi*((x(idx)-(t-1)/2)/t))).^2;
% case 2
idx = x>(t-1)/2 & x<=(1-t)/2;
f(idx) = 1;
% case 3
idx = x>(1-t)/2 & x<=(1+t)/2;
f(idx) = -(cos(0.5*pi*((x(idx)-(1+t)/2)/t))).^2+1;
end