how to plot this function?
Infos
Cette question est clôturée. Rouvrir pour modifier ou répondre.
Afficher commentaires plus anciens
if true
% code
endfunction [finalW, MSE, Numiterations] = gradientDescentLinearNeuron(P, t, lr, ErrorGoal, MaxIterations)
FinalW=(2*rand(1,100)-1)/10;
x=(2*rand(1,100)-1)/10;
x(:);
ErrorGoal=0.15;
lr=0.1;
step=1;
while ((MSE>ErrorGoal)&&(step<=MaxIterations))
y1=FinalW*x;
si=size(FinalW);
y2=ones(si);
y=[y1,y2];
dMSEdw=-1/((length(x)*((sum(t)-y)*x)));
FinalW=FinalW-lr*dMSEdw;
MSE=1/2*(length(x)*((sum(t)-y).^2));
step=step+1;
end
end
3 commentaires
madhan ravi
le 1 Nov 2018
P, t, lr, ErrorGoal, MaxIterations??
ariellewin11 ariellewin11
le 1 Nov 2018
Modifié(e) : ariellewin11 ariellewin11
le 1 Nov 2018
Kevin Chng
le 1 Nov 2018
function [finalW, MSE, Numiterations] = gradientDescentLinearNeuron(P, t, lr, ErrorGoal, MaxIterations)
FinalW=(2*rand(1,100)-1)/10;
x=(2*rand(1,100)-1)/10;
x(:);
ErrorGoal=0.15;
lr=0.1;
step=1;
while ((MSE>ErrorGoal)&&(step<=MaxIterations))
y1=FinalW*x;
si=size(FinalW);
y2=ones(si);
y=[y1,y2];
dMSEdw=-1/((length(x)*((sum(t)-y)*x)));
FinalW=FinalW-lr*dMSEdw;
MSE=1/2*(length(x)*((sum(t)-y).^2));
step=step+1;
end
end
What do you want to plot?
plot(Numiterations,finalW)
What is the error?
Réponses (0)
Cette question est clôturée.
Community Treasure Hunt
Find the treasures in MATLAB Central and discover how the community can help you!
Start Hunting!