linear regression , Grid search

I try to solve this linear system then to tray to do the same thing with Grid search method

function Gred_Search5()

clear

clc

close all

x=[1 ;2 ;3 ;4 ;5]; % die Matrix der Angaben am X Achse

y=[1 ;1 ;4 ;3 ;6]; % die Matrix der Angaben am y Achse

plot(x,y,'bo')

xlim([0 8])

ylim([0 8])

hold on

%%

step_size= 0.05 ;

x0=0 ;

while step_size > 0.01

%% these are three sample points I'd like to check

xsamples = [x0-step_size x0 x0+step_size];

%% these is the function evaluated at my samp,e points

fsamples = f(xsamples);

%%% this is the minimum of the tree

fmin =min(fsamples);

%%% this is the vector location of my minimum

loc = (fmin==fsamples) ; %%% this is a vector of 0's and 1's

x0= xsamples(loc); %%% Grabbing the x that computs fmin = f(x)

plot (x0 , f(x0), 'r*','MarkerSize',10)

pause(1.0)

end

end

function P_pre=f(x,y)

N=length(x);

G =[ones(N,1),x];

d_obs=y;

Gtr=G' ;

m_est= inv(G'* G)* G'*d_obs;

disp(m_est);

P_pre= G * m_est;

end

how I can fix this code?

Answer 1

[Sulaymon Eshkabilov]

There are a few errs that need to be fixed, see the given comments:

function Gred_Search5()
clear 
clc
close all 
x=[1 ;2 ;3 ;4 ;5]; % die Matrix der Angaben am X Achse
y=[1 ;1 ;4 ;3 ;6];  % die Matrix der Angaben am y Achse
plot(x,y,'bo')
xlim([0 8])
ylim([0 8])
hold on 
%%
step_size= 0.05 ;
x0=0 ;
while step_size > 0.01                                    % ERR: while loop has to be fixed. step_size value to be recomputed inside the loop
    %% these are three sample points I'd like to check
    xsamples = [x0-step_size x0 x0+step_size];            % ERR: size of xsamples is 3 while size of y is 5 as defined above
    %% these is the function evaluated at my samp,e points
    fsamples = f(xsamples);                               % ERR: f(x, y) requires two inputs
    %%% this is the minimum of the tree 
    fmin =min(fsamples);                                  % ERR:
    %%% this is the vector location of my minimum
    loc = (fmin==fsamples) ; %%% this is a  vector of 0's and 1's   % ERR:
    x0= xsamples(loc); %%% Grabbing the x that computs fmin = f(x)
    plot (x0 , f(x0), 'r*','MarkerSize',10)                         % ERR: f(x, y)
    pause(1.0)
    end
end
function P_pre=f(x,y)                                               % NOTE: two inputs are required
N=length(x);
G =[ones(N,1),x];                                                
d_obs=y;
Gtr=G' ;
m_est= inv(G'* G)* G'*d_obs;
disp(m_est);
P_pre= G * m_est;
end