Using Autoencoder to get the features from DWT decomposed coefficients

I have performed DWT decomposition using Daubechies wavelet on filtered ECG signal. Can anyone help me how to write the code for getting the features from these coefficients using autoencoder?

Here is my code. Thanks in advance.

clc
clearvars
close all
clc;
clearvars
lt=64;
btt=50;
rn1=0.009;
rn2=0.09;
data= importdata('filtered.csv');
x = data(:, 1);
y = data(:, 1);
tg=[];
Ya=y;
clear x y
V=[];
F3=[];
dt=[];
t1=[];
T4=[];
X3=[];
F1=[];
%%wavelet decomposition
clear i j k i1 i2 
Ya1=Ya;
clear Ya
Ya=Ya1((45:236+(256*btt)),:);
n= length (Ya);
n1= n/lt;
for i= 1:1:n1
    for j=1:1:lt
        k=lt*(i-1)+j;
        x2(j)=Ya(k);
    end
    [c,l]=wavedec(x2,5,'db4');
    [cd1,cd2,cd3,cd4,cd5] = detcoef(c,l,[1 2 3 4 5]);
    A=appcoef(c,l,'db4');
    ec1=((sum(cd1.^2))/(length(cd1)));
    ec2=((sum(cd2.^2))/(length(cd2)));
    ec3=((sum(cd3.^2))/(length(cd3)));
    ec4=((sum(cd4.^2))/(length(cd4)));
    ec5=((sum(cd5.^2))/(length(cd5)));
    ec6=((sum(A.^2))/(length(A)));
    dc=[ec1 ec2 ec3 ec4 ec5 ec6];
    dc1=sum(dc);
    dc=100.*((dc)/dc1);
    df=dc(:,(1:3));
    xm=mean(x2);
    x21=x2-xm;
    xe1=(((sum(x21.^2))/lt).^0.5);
    xe2=100.*((((sum((cd1-mean(cd1)).^2))/lt).^0.5)/xe1);
    xe3=100.*((((sum((cd2-mean(cd2)).^2))/lt).^0.5)/xe1);
    xe4=100.*((((sum((cd3-mean(cd3)).^2))/lt).^0.5)/xe1);
    rt=xe2+xe3+xe4;
   
    if xe1<=rn1
        upt=4;
    elseif xe1>rn1 && xe1<rn2
        upt=4;
    elseif xe1>=rn2
        upt=2;
    end
    

          
    tg=[tg;xe1];
for i1=1:1:upt
    g(i1)=l((length(l)-1)-upt+i1);
end
if upt>0
g1=sum(g);
else
    g1=0;
end
if g1>0
for i2=length(c)-g1+1:length(c)
    c(i2)=0;
end 
end
if xe1<=0.05
    ty=[6 7];
elseif xe1>0.05 && xe1<.2
    ty=[5 6];
elseif xe1>=0.2
    ty=[4 5];
end
upt;
[x,T3]=coding(upt,x2, c,l,0,ty,i);
T4=[T4;T3];
x=[i 0 0 0 0; x]; 
[F2]=recon(upt,c,l,x);
if i==0
    x
end
[F]=waverec(c,l,'db4');
[prd1]=calprd(x2,F);
df=[df xe1 prd1 rt];
dt=[dt;df];
F1=[F1 F];
X3=[X3 x2];
F3=[F3 F2];
t1=[t1 mean((diff(X3)).^2)];
    clear dc dc1 dc g g1 c F x2 c l A F2
end

Answer 1

[ariTU]

Did you come up with a solution?

I would be interested in your solution. Thanks :)