I need to convert this Matlab code to Python code.
You can see my attempt below . but when i run the python code get error message ShapeError: Matrix size mismatch: (3, 6) * (1296, 1296).

Here is the Matlab code.

clc; clear all;
A11 = 83.5165
B11 = 0.0000
D11 = 0.2784

H =24.3590
I0 = 792.0000
I1 = 0.0000
I2 = 2.6400
fac=1e-3 % Initial value
kb=0.;err=+1;ico=0;isign=1;
while err>0
ico=ico+1;  kb=kb+fac; od=kb^2;
c1=vpa((D11*I0 - B11*I1)*od^2/(B11^2 - A11*D11),10);
c2=vpa((B11*H + (D11*I1 - B11*I2)*od^2)/(B11^2 - A11*D11),10);
c3=vpa(B11*H/(B11^2 - A11*D11),10);
c4=vpa(-(B11*I0 - A11*I1)*od^2/(B11^2 - A11*D11),10) ;
c5=vpa(-(A11*H + (B11*I1 - A11*I2)*od^2)/(B11^2 - A11*D11),10);
c6=vpa(-A11*H/(B11^2 - A11*D11),10);
c7=vpa(-H/(H-N0),10);
c8=vpa(-I0*od^2/(H-N0),10);
T = [0   1   0   0   0    0   ;...
     eval(c1)  0   eval(c2)  0   0    eval(c3) ;...
     0   0   0   1   0    0;...
     eval(c4)  0   eval(c5)  0   0    eval(c6)  ;...
     0   0   0   0   0    1 ;...
     0   0   0   eval(c7)  eval(c8)   0];
    [E,lambda]=eig(T); lambda=diag(lambda);
    gdet1= E;
    gdet2= E;
    
    tt1 =eval(A11)*gdet1(2,:)+eval(B11)*gdet1(4,:);
    tt2 =gdet1(5,:);
    tt3 =eval(B11)*gdet1(2,:)+eval(D11)*gdet1(4,:);
    tt4 =eval(A11)*gdet2(2,:)+eval(B11)*gdet2(4,:);
    tt5 =gdet2(5,:);
    tt6 =eval(B11)*gdet2(2,:)+eval(D11)*gdet2(4,:);
    
    tem1=[tt1;tt2;tt3];
    tem2=[tt4;tt5;tt6];
    
    lam1= lambda*1/2;
    lam2= lambda*(-1/2);
    e_lam1 = diag(exp(lam1),0); %e_lam1(1,2) = a/2;
    e_lam2 = diag(exp(lam2),0); %e_lam2(1,2) = -a/2;
   % print(lam1,lam2);
    bc1 =tem1*e_lam1;
    bc2 =tem2*e_lam2;
    bc  =[bc1;bc2];
    for ii=1:6
        if real(lam1(ii))>100 % large positive
            bc(:,ii)=exp(imag(lam1(ii))*1i)*[tt1(ii); tt2(ii); tt3(ii);0;0;0]';
        elseif real(lam1(ii))<-100 % large negative
            bc(:,ii)=exp(imag(lam1(ii))*1i)*[0;0;0;tt4(ii); tt5(ii); tt6(ii)]';
        end
    end
        err=det(bc)/det(E);
    %% To avoid looping stops at once with just 1 loop
    if ico==1
        if err<0
            isign=-1;
        else
            isign=1;
        end
    end
err=err*isign;
    %% To decide the tolerance - if the error is quite large when err<0, then err =-err to loop again
    if err<0
        kb=kb-fac;
        fac=fac/10;
        if fac>1.0e-5
            err=-err;
        end
    end
fprintf('er= %e | fac= %e| od: %4.4f n',err, fac, od);

Here is my attempt, the Python code

import sympy as sp


A11 = 83.5165
B11 = 0.0000
D11 = 0.2784

H =24.3590
I0 = 792.0000
I1 = 0.0000
I2 = 2.6400
fac = 1e-3  # Initial value
kb = 0
err = 1
ico = 0
isign = 1

while err > 0:
    ico += 1
    kb += fac
    od = kb**2
    c1 = sp.N((D11 * I0 - B11 * I1) * od**2 / (B11**2 - A11 * D11), 10)
    c2 = sp.N((B11 * H + (D11 * I1 - B11 * I2) * od**2) / (B11**2 - A11 * D11), 10)
    c3 = sp.N(B11 * H / (B11**2 - A11 * D11), 10)
    c4 = sp.N(-(B11 * I0 - A11 * I1) * od**2 / (B11**2 - A11 * D11), 10)
    c5 = sp.N(-(A11 * H + (B11 * I1 - A11 * I2) * od**2) / (B11**2 - A11 * D11), 10)
    c6 = sp.N(-A11 * H / (B11**2 - A11 * D11), 10)
    c7 = sp.N(-H / (H), 10)
    c8 = sp.N(-I0 * od**2 / (H), 10)
    T = sp.Matrix([
        [0, 1, 0, 0, 0, 0],
        [c1, 0, c2, 0, 0, c3],
        [0, 0, 0, 1, 0, 0],
        [c4, 0, c5, 0, 0, c6],
        [0, 0, 0, 0, 0, 1],
        [0, 0, 0, c7, c8, 0]
    ])

    E, lambda_ = T.diagonalize() 
    lambda_ = sp.diag(*lambda_) 
    
    gdet1 = E  
    gdet2 = E 

    tt1 = A11 * gdet1[1, :] + B11 * gdet1[3, :]
    tt2 = gdet1[4, :]  
    tt3 = B11 * gdet1[1, :] + D11 * gdet1[3, :]  
    tt4 = A11 * gdet2[1, :] + B11 * gdet2[3, :] 
    tt5 = gdet2[4, :] 
    tt6 = B11 * gdet2[1, :] + D11 * gdet2[3, :] 

    tem1 = sp.Matrix([tt1, tt2, tt3]) 
    tem2 = sp.Matrix([tt4, tt5, tt6]) 

    lam1 = lambda_ / 2 
    lam2 = -lambda_ / 2 
    e_lam1 = sp.diag(*[sp.exp(l) for l in lam1]) 
    e_lam2 = sp.diag(*[sp.exp(l) for l in lam2]) 

    bc1 = tem1 * e_lam1 
    bc2 = tem2 * e_lam2 
    bc = sp.Matrix.vstack(bc1, bc2) 

    for ii in range(6): 
        if sp.re(lam1[ii]) > 100:  
            bc[:, ii] = sp.exp(sp.im(lam1[ii]) * sp.I) * sp.Matrix([tt1[ii], tt2[ii], tt3[ii], 0, 0, 0]) 
        elif sp.re(lam1[ii]) < -100:  # in matlab  elseif real(lam1(ii))<-100 
            bc[:, ii] = sp.exp(sp.im(lam1[ii]) * sp.I) * sp.Matrix([0, 0, 0, tt4[ii], tt5[ii], tt6[ii]]) 

    err = sp.det(bc) / sp.det(E) 

    
    if ico == 1: 
        if err < 0: 
            isign = -1 
        else:  
            isign = 1

    err *= isign

    if err < 0:
        kb -= fac
        fac /= 10
        if fac > 1.0e-5:
            err = -err
    print(f'er= {err:e} | fac= {fac:e} | od: {od:.4f} ')

The Python code isn’t functioning properly. Can someone help me correct it?

Thanks in advance.