% Argyris Zymnis, Joelle Skaf, Stephen Boyd
%
% We are given a matrix A in R^{m*n}
% and are interested in solving the problem:
%
% minimize    ||A - Y*X||_F
% subject to  Y >= 0, X >= 0
%
% where Y in R{m*k} and X in R{k*n}.
% This script generates a random matrix A and obtains an
% *approximate* solution to the above problem by first generating
% a random initial guess for Y and the alternatively minimizing
% over X and Y for a fixed number of iterations.

% Generate data matrix A
rstate = rand('state');
m = 10; n = 10; k = 5;
A = rand(m,k)*rand(k,n);

% Initialize Y randomly
Y = rand(m,k);

% Perform alternating minimization
MAX_ITERS = 30;
residual = zeros(1,MAX_ITERS);
for iter = 1:MAX_ITERS
    cvx_begin
        if mod(iter,2) == 1
            variable X(k,n)
            X >= 0;
        else
            variable Y(m,k)
            Y >= 0;
        end
        minimize(norm(A - Y*X,'fro'));
    cvx_end
    fprintf(1,'Iteration %d, residual norm %g\n',iter,cvx_optval);
    residual(iter) = cvx_optval;
end

% Plot residuals
plot(residual);
xlabel('Iteration Number');
ylabel('Residual Norm');

% Display results
disp( 'Original matrix:' );
disp( A );
disp( 'Left factor Y:' );
disp( Y );
disp( 'Right factor X:' );
disp( X );
disp( 'Residual A - Y * X:' );
disp( A - Y * X );
fprintf( 'Residual after %d iterations: %g\n', iter, cvx_optval );
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :  -2.75E+00 2.70E-01 0.000 0.7218 0.9000 0.9000   3.54  1  1  2.2E+00
  2 :  -3.18E+00 1.08E-01 0.000 0.3995 0.9000 0.9000   3.83  1  1  3.4E-01
  3 :  -3.41E+00 2.55E-02 0.000 0.2361 0.9000 0.9000   1.61  1  1  6.2E-02
  4 :  -3.43E+00 8.23E-03 0.000 0.3229 0.9000 0.9000   1.18  1  1  1.9E-02
  5 :  -3.42E+00 3.03E-03 0.000 0.3677 0.9000 0.9000   1.07  1  1  6.7E-03
  6 :  -3.42E+00 9.22E-04 0.000 0.3045 0.9000 0.9000   1.03  1  1  2.0E-03
  7 :  -3.42E+00 2.09E-04 0.000 0.2263 0.0000 0.9000   1.01  1  1  1.2E-03
  8 :  -3.42E+00 6.03E-05 0.000 0.2890 0.9000 0.9155   1.01  1  1  3.4E-04
  9 :  -3.42E+00 5.20E-06 0.000 0.0863 0.6803 0.9000   1.00  1  1  1.0E-04
 10 :  -3.42E+00 3.36E-07 0.000 0.0646 0.9900 0.9900   1.00  1  1  6.7E-06
 11 :  -3.42E+00 8.94E-09 0.000 0.0266 0.9900 0.9850   1.00  1  1  1.9E-07
 12 :  -3.42E+00 9.29E-10 0.000 0.1039 0.9060 0.9000   1.00  1  1  2.1E-08
 13 :  -3.42E+00 2.50E-10 0.103 0.2686 0.7374 0.9000   1.00  1  1  5.6E-09

iter seconds digits       c*x               b*y
 13      0.1   Inf -3.4206050476e+00 -3.4206050447e+00
|Ax-b| =   4.0e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  5.0e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    7.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +3.42061
Iteration 1, residual norm 3.42061
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   5.61E-02 2.27E-01 0.000 0.6069 0.9000 0.9000   5.23  1  1  1.0E+00
  2 :  -6.49E-01 5.81E-02 0.000 0.2557 0.9000 0.9000   2.12  1  1  1.7E-01
  3 :  -5.10E-01 1.75E-02 0.000 0.3012 0.9000 0.9000   1.20  1  1  4.7E-02
  4 :  -4.65E-01 5.59E-03 0.000 0.3193 0.9000 0.9000   1.13  1  1  1.4E-02
  5 :  -4.55E-01 1.89E-03 0.000 0.3375 0.9000 0.9000   1.07  1  1  4.7E-03
  6 :  -4.50E-01 5.37E-04 0.000 0.2849 0.9000 0.9000   1.03  1  1  1.5E-03
  7 :  -4.49E-01 4.63E-05 0.000 0.0861 0.9000 0.9085   1.01  1  1  2.7E-04
  8 :  -4.48E-01 9.33E-07 0.000 0.0202 0.9336 0.9900   1.00  1  1  1.7E-05
  9 :  -4.48E-01 7.13E-08 0.438 0.0764 0.9900 0.9900   1.00  1  1  1.3E-06
 10 :  -4.48E-01 2.73E-09 0.000 0.0382 0.9737 0.9900   1.00  1  1  4.7E-08
 11 :  -4.48E-01 1.95E-10 0.472 0.0715 0.9900 0.9900   1.00  1  1  3.4E-09

iter seconds digits       c*x               b*y
 11      0.1   Inf -4.4806595221e-01 -4.4806595094e-01
|Ax-b| =   1.5e-08, [Ay-c]_+ =   1.6E-09, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    6.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 2.1531.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.448066
Iteration 2, residual norm 0.448066
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   1.71E-02 2.11E-01 0.000 0.5621 0.9000 0.9000   5.35  1  1  8.8E-01
  2 :  -1.78E-01 6.84E-02 0.000 0.3250 0.9000 0.9000   2.14  1  1  1.8E-01
  3 :  -2.08E-01 1.99E-02 0.000 0.2911 0.9000 0.9000   1.48  1  1  4.3E-02
  4 :  -2.23E-01 5.39E-03 0.000 0.2707 0.9000 0.9000   1.20  1  1  1.1E-02
  5 :  -2.26E-01 1.33E-03 0.000 0.2462 0.9000 0.9000   1.08  1  1  3.5E-03
  6 :  -2.26E-01 2.95E-04 0.000 0.2226 0.9000 0.9000   1.03  1  1  8.6E-04
  7 :  -2.26E-01 5.79E-05 0.000 0.1960 0.8986 0.9000   1.01  1  1  1.5E-04
  8 :  -2.26E-01 1.08E-05 0.000 0.1869 0.8678 0.9000   1.00  1  1  2.9E-05
  9 :  -2.26E-01 1.14E-06 0.000 0.1052 0.9450 0.9451   1.00  1  1  3.1E-06
 10 :  -2.26E-01 1.81E-07 0.143 0.1586 0.9000 0.9044   1.00  1  1  5.4E-07
 11 :  -2.26E-01 2.67E-08 0.253 0.1481 0.9000 0.9099   1.00  1  1  9.8E-08
 12 :  -2.26E-01 4.16E-09 0.194 0.1556 0.9000 0.9080   1.00  1  1  1.7E-08
 13 :  -2.26E-01 6.58E-10 0.159 0.1581 0.9000 0.9021   1.00  1  1  2.8E-09

iter seconds digits       c*x               b*y
 13      0.1   8.0 -2.2594586505e-01 -2.2594586721e-01
|Ax-b| =   7.2e-09, [Ay-c]_+ =   4.6E-10, |x|=  1.4e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    9.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.08001.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.225946
Iteration 3, residual norm 0.225946
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   6.71E-02 2.25E-01 0.000 0.6009 0.9000 0.9000   5.27  1  1  1.0E+00
  2 :  -4.20E-01 6.53E-02 0.000 0.2902 0.9000 0.9000   2.10  1  1  1.9E-01
  3 :  -2.12E-01 2.23E-02 0.000 0.3418 0.9000 0.9000   1.23  1  1  6.0E-02
  4 :  -1.53E-01 6.66E-03 0.000 0.2985 0.9000 0.9000   1.13  1  1  1.7E-02
  5 :  -1.44E-01 2.34E-03 0.000 0.3518 0.9000 0.9000   1.05  1  1  5.9E-03
  6 :  -1.39E-01 1.87E-04 0.000 0.0796 0.0000 0.9000   1.02  1  1  3.5E-03
  7 :  -1.41E-01 4.48E-05 0.000 0.2398 0.9000 0.9109   1.01  1  1  9.2E-04
  8 :  -1.37E-01 2.39E-06 0.000 0.0533 0.9864 0.9900   1.00  1  1  4.9E-05
  9 :  -1.37E-01 2.19E-07 0.267 0.0916 0.9900 0.9900   1.00  1  1  4.5E-06
 10 :  -1.37E-01 6.77E-08 0.000 0.3096 0.9000 0.9371   1.00  1  1  1.5E-06
 11 :  -1.37E-01 4.21E-09 0.194 0.0622 0.9902 0.9900   1.00  1  1  9.4E-08
 12 :  -1.37E-01 6.11E-10 0.000 0.1452 0.9088 0.9000   1.00  1  1  1.4E-08

iter seconds digits       c*x               b*y
 12      0.1   Inf -1.3710121956e-01 -1.3710119365e-01
|Ax-b| =   9.1e-08, [Ay-c]_+ =   3.3E-10, |x|=  1.4e+00, |y|=  5.0e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    7.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 2.01709.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.137101
Iteration 4, residual norm 0.137101
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   5.12E-02 2.11E-01 0.000 0.5629 0.9000 0.9000   5.37  1  1  8.7E-01
  2 :  -2.20E-02 6.17E-02 0.000 0.2928 0.9000 0.9000   2.12  1  1  1.6E-01
  3 :  -8.25E-02 2.25E-02 0.000 0.3637 0.9000 0.9000   1.43  1  1  5.0E-02
  4 :  -9.07E-02 5.88E-03 0.000 0.2620 0.9000 0.9000   1.20  1  1  1.2E-02
  5 :  -9.61E-02 4.55E-04 0.380 0.0774 0.9900 0.9900   1.09  1  1  2.1E-03
  6 :  -9.61E-02 1.78E-04 0.000 0.3907 0.9000 0.9000   1.02  1  1  8.4E-04
  7 :  -9.60E-02 3.84E-05 0.000 0.2158 0.8148 0.9000   1.01  1  1  1.8E-04
  8 :  -9.60E-02 7.98E-06 0.000 0.2079 0.9019 0.9000   1.00  1  1  3.5E-05
  9 :  -9.60E-02 1.62E-06 0.029 0.2032 0.9012 0.9000   1.00  1  1  7.2E-06
 10 :  -9.60E-02 3.23E-07 0.125 0.1989 0.9009 0.9000   1.00  1  1  1.4E-06
 11 :  -9.60E-02 6.50E-08 0.178 0.2014 0.9009 0.9000   1.00  1  1  2.8E-07
 12 :  -9.60E-02 1.32E-08 0.207 0.2030 0.9005 0.9000   1.00  1  1  5.8E-08
 13 :  -9.60E-02 2.70E-09 0.229 0.2047 0.9000 0.9001   1.00  1  1  1.2E-08

iter seconds digits       c*x               b*y
 13      0.1   7.3 -9.6013840647e-02 -9.6013845288e-02
|Ax-b| =   1.1e-08, [Ay-c]_+ =   3.7E-09, |x|=  1.5e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    8.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0960138
Iteration 5, residual norm 0.0960138
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.24E-02 2.25E-01 0.000 0.6011 0.9000 0.9000   5.28  1  1  1.0E+00
  2 :  -3.70E-01 6.53E-02 0.000 0.2901 0.9000 0.9000   2.09  1  1  1.9E-01
  3 :  -1.53E-01 2.19E-02 0.000 0.3351 0.9000 0.9000   1.23  1  1  5.9E-02
  4 :  -8.59E-02 6.70E-03 0.000 0.3061 0.9000 0.9000   1.12  1  1  1.7E-02
  5 :  -7.78E-02 2.07E-03 0.000 0.3095 0.9000 0.9000   1.05  1  1  5.3E-03
  6 :  -7.31E-02 1.54E-04 0.000 0.0745 0.0000 0.9000   1.02  1  1  3.0E-03
  7 :  -7.40E-02 3.10E-05 0.000 0.2005 0.9000 0.9067   1.01  1  1  6.4E-04
  8 :  -7.19E-02 1.28E-06 0.000 0.0415 0.9901 0.9900   1.00  1  1  2.6E-05
  9 :  -7.19E-02 4.90E-08 0.000 0.0382 0.9900 0.9904   1.00  1  1  1.1E-06
 10 :  -7.19E-02 1.62E-08 0.261 0.3310 0.9000 0.9252   1.00  1  1  3.5E-07
 11 :  -7.19E-02 8.74E-10 0.289 0.0539 0.9900 0.7357   1.00  1  1  2.2E-08
 12 :  -7.19E-02 1.73E-11 0.000 0.0198 0.9827 0.9900   1.00  1  1  4.3E-10

iter seconds digits       c*x               b*y
 12      0.1   Inf -7.1869162683e-02 -7.1869162321e-02
|Ax-b| =   2.7e-09, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  5.0e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    7.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.9701.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0718692
Iteration 6, residual norm 0.0718692
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   6.05E-02 2.13E-01 0.000 0.5689 0.9000 0.9000   5.38  1  1  8.9E-01
  2 :   9.92E-03 6.28E-02 0.000 0.2948 0.9000 0.9000   2.13  1  1  1.6E-01
  3 :  -4.25E-02 2.11E-02 0.000 0.3359 0.9000 0.9000   1.42  1  1  4.7E-02
  4 :  -5.35E-02 4.85E-03 0.000 0.2298 0.9000 0.9000   1.20  1  1  9.9E-03
  5 :  -5.79E-02 4.60E-04 0.249 0.0948 0.9900 0.9900   1.07  1  1  2.1E-03
  6 :  -5.82E-02 1.77E-04 0.000 0.3844 0.9000 0.9000   1.02  1  1  9.0E-04
  7 :  -5.81E-02 4.58E-05 0.000 0.2590 0.7998 0.9000   1.01  1  1  2.6E-04
  8 :  -5.81E-02 1.11E-05 0.000 0.2433 0.9000 0.9008   1.00  1  1  6.6E-05
  9 :  -5.82E-02 2.71E-06 0.000 0.2433 0.9000 0.9050   1.00  1  1  1.9E-05
 10 :  -5.82E-02 6.09E-07 0.000 0.2249 0.9000 0.9095   1.00  1  1  5.2E-06
 11 :  -5.82E-02 1.11E-07 0.000 0.1815 0.9000 0.9098   1.00  1  1  1.1E-06
 12 :  -5.82E-02 1.63E-08 0.063 0.1477 0.9000 0.9048   1.00  1  1  1.8E-07
 13 :  -5.82E-02 1.58E-09 0.380 0.0966 0.9450 0.9488   1.00  1  1  2.0E-08
 14 :  -5.82E-02 1.54E-10 0.381 0.0977 0.9450 0.9489   1.00  1  1  2.1E-09

iter seconds digits       c*x               b*y
 14      0.1   7.9 -5.8151848773e-02 -5.8151849456e-02
|Ax-b| =   2.6e-09, [Ay-c]_+ =   1.4E-11, |x|=  1.5e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    8.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.01059.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0581518
Iteration 7, residual norm 0.0581518
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.35E-02 2.25E-01 0.000 0.6012 0.9000 0.9000   5.29  1  1  1.0E+00
  2 :  -3.45E-01 6.70E-02 0.000 0.2977 0.9000 0.9000   2.08  1  1  2.0E-01
  3 :  -1.28E-01 2.22E-02 0.000 0.3311 0.9000 0.9000   1.24  1  1  6.0E-02
  4 :  -6.14E-02 6.96E-03 0.000 0.3138 0.9000 0.9000   1.12  1  1  1.8E-02
  5 :  -5.44E-02 1.93E-03 0.000 0.2770 0.9000 0.9000   1.05  1  1  4.9E-03
  6 :  -5.07E-02 5.21E-04 0.000 0.2702 0.9000 0.9000   1.02  1  1  1.4E-03
  7 :  -5.00E-02 8.14E-05 0.000 0.1561 0.9000 0.9052   1.01  1  1  6.2E-04
  8 :  -4.99E-02 8.53E-06 0.000 0.1049 0.9000 0.9104   1.00  1  1  1.2E-04
  9 :  -4.98E-02 3.84E-07 0.268 0.0450 0.8910 0.9900   1.00  1  1  8.0E-06
 10 :  -4.98E-02 6.16E-08 0.000 0.1606 0.9000 0.9141   1.00  1  1  1.5E-06
 11 :  -4.98E-02 1.00E-08 0.172 0.1624 0.9000 0.9058   1.00  1  1  2.4E-07
 12 :  -4.98E-02 1.93E-09 0.296 0.1931 0.9000 0.9151   1.00  1  1  4.9E-08
 13 :  -4.98E-02 3.44E-10 0.229 0.1779 0.9000 0.9125   1.00  1  1  8.6E-09

iter seconds digits       c*x               b*y
 13      0.1   Inf -4.9805568568e-02 -4.9805550094e-02
|Ax-b| =   6.1e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    7.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.94796.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0498056
Iteration 8, residual norm 0.0498056
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   6.33E-02 2.14E-01 0.000 0.5723 0.9000 0.9000   5.39  1  1  9.0E-01
  2 :   2.03E-02 6.38E-02 0.000 0.2979 0.9000 0.9000   2.14  1  1  1.7E-01
  3 :  -2.87E-02 1.48E-02 0.000 0.2326 0.9000 0.9000   1.43  1  1  3.2E-02
  4 :  -3.97E-02 3.27E-03 0.000 0.2200 0.9000 0.9000   1.17  1  1  6.6E-03
  5 :  -4.24E-02 7.35E-04 0.000 0.2250 0.9000 0.9000   1.06  1  1  1.5E-03
  6 :  -4.31E-02 1.70E-04 0.000 0.2318 0.9000 0.9000   1.02  1  1  7.4E-04
  7 :  -4.33E-02 3.08E-05 0.000 0.1808 0.9000 0.9076   1.01  1  1  2.7E-04
  8 :  -4.33E-02 6.05E-06 0.000 0.1963 0.9000 0.9089   1.00  1  1  7.0E-05
  9 :  -4.33E-02 1.38E-06 0.000 0.2288 0.9000 0.9131   1.00  1  1  1.9E-05
 10 :  -4.33E-02 3.35E-07 0.000 0.2419 0.9000 0.9148   1.00  1  1  4.9E-06
 11 :  -4.33E-02 7.40E-08 0.000 0.2212 0.9000 0.9104   1.00  1  1  1.1E-06
 12 :  -4.33E-02 6.57E-09 0.492 0.0888 0.9901 0.9900   1.00  1  1  9.2E-08
 13 :  -4.33E-02 1.42E-09 0.000 0.2163 0.9076 0.9000   1.00  1  1  1.5E-08

iter seconds digits       c*x               b*y
 13      0.1   7.1 -4.3347200036e-02 -4.3347203921e-02
|Ax-b| =   8.0e-09, [Ay-c]_+ =   2.0E-10, |x|=  1.5e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    8.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0433472
Iteration 9, residual norm 0.0433472
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.35E-02 2.25E-01 0.000 0.6013 0.9000 0.9000   5.30  1  1  1.0E+00
  2 :  -3.31E-01 6.79E-02 0.000 0.3016 0.9000 0.9000   2.08  1  1  2.0E-01
  3 :  -1.14E-01 2.25E-02 0.000 0.3306 0.9000 0.9000   1.24  1  1  6.1E-02
  4 :  -4.85E-02 7.14E-03 0.000 0.3179 0.9000 0.9000   1.12  1  1  1.9E-02
  5 :  -4.19E-02 1.88E-03 0.000 0.2636 0.9000 0.9000   1.05  1  1  4.8E-03
  6 :  -3.90E-02 5.24E-04 0.000 0.2786 0.9000 0.9000   1.02  1  1  1.6E-03
  7 :  -3.85E-02 6.78E-05 0.000 0.1293 0.9000 0.9078   1.01  1  1  8.1E-04
  8 :  -3.84E-02 7.14E-06 0.000 0.1054 0.9000 0.9095   1.00  1  1  1.4E-04
  9 :  -3.84E-02 4.97E-07 0.378 0.0695 0.9900 0.9900   1.00  1  1  9.2E-06
 10 :  -3.84E-02 3.65E-09 0.000 0.0074 0.9900 0.9903   1.00  1  1  9.0E-08
 11 :  -3.84E-02 1.25E-10 0.321 0.0342 0.9677 0.9900   1.00  1  1  3.1E-09

iter seconds digits       c*x               b*y
 11      0.1   Inf -3.8351836743e-02 -3.8351836060e-02
|Ax-b| =   2.0e-08, [Ay-c]_+ =   3.2E-11, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    6.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.93754.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0383518
Iteration 10, residual norm 0.0383518
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   6.50E-02 2.15E-01 0.000 0.5744 0.9000 0.9000   5.40  1  1  9.0E-01
  2 :   2.59E-02 6.46E-02 0.000 0.3005 0.9000 0.9000   2.14  1  1  1.7E-01
  3 :  -1.69E-02 1.27E-02 0.000 0.1970 0.9000 0.9000   1.43  1  1  2.7E-02
  4 :  -3.01E-02 3.28E-03 0.000 0.2572 0.9000 0.9000   1.16  1  1  6.6E-03
  5 :  -3.30E-02 7.70E-04 0.000 0.2349 0.9000 0.9000   1.06  1  1  1.5E-03
  6 :  -3.39E-02 1.71E-04 0.000 0.2223 0.9000 0.9000   1.02  1  1  5.7E-04
  7 :  -3.41E-02 2.60E-05 0.000 0.1522 0.9000 0.9050   1.01  1  1  2.2E-04
  8 :  -3.41E-02 2.88E-06 0.016 0.1105 0.9450 0.9459   1.00  1  1  2.8E-05
  9 :  -3.41E-02 1.24E-07 0.404 0.0433 0.9901 0.9900   1.00  1  1  1.4E-06
 10 :  -3.41E-02 2.47E-08 0.000 0.1986 0.9000 0.9022   1.00  1  1  3.1E-07
 11 :  -3.41E-02 1.86E-09 0.424 0.0754 0.9900 0.9900   1.00  1  1  2.4E-08
 12 :  -3.41E-02 4.47E-10 0.343 0.2401 0.9000 0.9140   1.00  1  1  7.0E-09

iter seconds digits       c*x               b*y
 12      0.1   7.4 -3.4135896358e-02 -3.4135897787e-02
|Ax-b| =   3.4e-09, [Ay-c]_+ =   2.5E-10, |x|=  1.5e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    6.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0341359
Iteration 11, residual norm 0.0341359
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.31E-02 2.25E-01 0.000 0.6014 0.9000 0.9000   5.30  1  1  1.0E+00
  2 :  -3.22E-01 6.85E-02 0.000 0.3043 0.9000 0.9000   2.08  1  1  2.0E-01
  3 :  -1.04E-01 2.26E-02 0.000 0.3304 0.9000 0.9000   1.24  1  1  6.2E-02
  4 :  -3.97E-02 7.22E-03 0.000 0.3187 0.9000 0.9000   1.12  1  1  1.9E-02
  5 :  -3.32E-02 1.80E-03 0.000 0.2490 0.9000 0.9000   1.06  1  1  4.6E-03
  6 :  -3.10E-02 5.07E-04 0.000 0.2820 0.9000 0.9000   1.02  1  1  2.1E-03
  7 :  -3.08E-02 7.14E-05 0.000 0.1409 0.9000 0.9088   1.01  1  1  1.2E-03
  8 :  -3.07E-02 8.02E-06 0.000 0.1124 0.9000 0.9130   1.00  1  1  2.2E-04
  9 :  -3.07E-02 3.73E-07 0.000 0.0465 0.9000 0.9141   1.00  1  1  8.8E-06
 10 :  -3.07E-02 1.29E-08 0.295 0.0345 0.9900 0.9885   1.00  1  1  3.3E-07
 11 :  -3.07E-02 2.30E-09 0.113 0.1784 0.9000 0.4012   1.00  1  1  6.3E-08
 12 :  -3.07E-02 5.72E-11 0.000 0.0249 0.9900 0.9900   1.00  1  1  1.6E-09

iter seconds digits       c*x               b*y
 12      0.1   Inf -3.0650643241e-02 -3.0650642739e-02
|Ax-b| =   1.1e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    7.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.92925.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0306506
Iteration 12, residual norm 0.0306506
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   6.63E-02 2.16E-01 0.000 0.5759 0.9000 0.9000   5.40  1  1  9.0E-01
  2 :   2.96E-02 6.53E-02 0.000 0.3030 0.9000 0.9000   2.14  1  1  1.7E-01
  3 :  -6.98E-03 1.42E-02 0.000 0.2176 0.9000 0.9000   1.43  1  1  3.1E-02
  4 :  -2.27E-02 3.79E-03 0.000 0.2664 0.9000 0.9000   1.17  1  1  7.6E-03
  5 :  -2.64E-02 8.24E-04 0.000 0.2176 0.9000 0.9000   1.06  1  1  1.6E-03
  6 :  -2.73E-02 1.82E-04 0.000 0.2205 0.9000 0.9000   1.02  1  1  3.5E-04
  7 :  -2.76E-02 1.45E-05 0.476 0.0799 0.9900 0.9900   1.01  1  1  8.7E-05
  8 :  -2.76E-02 1.12E-06 0.349 0.0771 0.9906 0.9900   1.00  1  1  2.0E-06
  9 :  -2.76E-02 2.38E-07 0.215 0.2129 0.9000 0.9110   1.00  1  1  1.3E-06
 10 :  -2.76E-02 5.42E-08 0.177 0.2273 0.9000 0.9083   1.00  1  1  5.1E-07
 11 :  -2.76E-02 1.17E-08 0.094 0.2155 0.9000 0.9028   1.00  1  1  1.3E-07
 12 :  -2.76E-02 2.24E-09 0.067 0.1917 0.9025 0.9000   1.00  1  1  2.4E-08
 13 :  -2.76E-02 4.31E-10 0.137 0.1923 0.9046 0.9000   1.00  1  1  4.1E-09

iter seconds digits       c*x               b*y
 13      0.1   7.7 -2.7613366413e-02 -2.7613367034e-02
|Ax-b| =   1.8e-09, [Ay-c]_+ =   5.6E-10, |x|=  1.5e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    9.000E-02    1.000E-02    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0276134
Iteration 13, residual norm 0.0276134
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.26E-02 2.25E-01 0.000 0.6015 0.9000 0.9000   5.30  1  1  1.0E+00
  2 :  -3.17E-01 6.90E-02 0.000 0.3062 0.9000 0.9000   2.08  1  1  2.1E-01
  3 :  -9.70E-02 2.28E-02 0.000 0.3303 0.9000 0.9000   1.25  1  1  6.2E-02
  4 :  -3.33E-02 7.24E-03 0.000 0.3179 0.9000 0.9000   1.12  1  1  1.9E-02
  5 :  -2.65E-02 1.64E-03 0.000 0.2260 0.9000 0.9000   1.06  1  1  4.5E-03
  6 :  -2.51E-02 4.50E-04 0.000 0.2748 0.9000 0.9000   1.02  1  1  3.0E-03
  7 :  -2.51E-02 6.81E-05 0.000 0.1514 0.9000 0.9100   1.01  1  1  1.5E-03
  8 :  -2.51E-02 9.83E-06 0.000 0.1444 0.9000 0.9163   1.01  1  1  3.7E-04
  9 :  -2.50E-02 1.20E-06 0.000 0.1219 0.9000 0.9167   1.00  1  1  5.0E-05
 10 :  -2.50E-02 8.13E-08 0.000 0.0679 0.9000 0.9207   1.00  1  1  2.2E-06
 11 :  -2.50E-02 7.31E-09 0.223 0.0899 0.9900 0.8680   1.00  1  1  2.0E-07
 12 :  -2.50E-02 2.85E-10 0.085 0.0389 0.9613 0.9900   1.00  1  1  7.7E-09

iter seconds digits       c*x               b*y
 12      0.1   Inf -2.5015523269e-02 -2.5015521652e-02
|Ax-b| =   5.5e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    6.000E-02    1.000E-02    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.92165.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0250155
Iteration 14, residual norm 0.0250155
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   6.75E-02 2.16E-01 0.000 0.5769 0.9000 0.9000   5.41  1  1  9.1E-01
  2 :   3.22E-02 6.59E-02 0.000 0.3052 0.9000 0.9000   2.14  1  1  1.7E-01
  3 :  -5.81E-04 1.49E-02 0.000 0.2256 0.9000 0.9000   1.44  1  1  3.2E-02
  4 :  -1.72E-02 4.03E-03 0.000 0.2706 0.9000 0.9000   1.17  1  1  8.1E-03
  5 :  -2.14E-02 8.33E-04 0.000 0.2069 0.9000 0.9000   1.06  1  1  1.6E-03
  6 :  -2.24E-02 1.78E-04 0.000 0.2140 0.9000 0.9000   1.02  1  1  3.5E-04
  7 :  -2.27E-02 2.74E-05 0.000 0.1539 0.9000 0.9017   1.01  1  1  1.0E-04
  8 :  -2.27E-02 4.80E-06 0.101 0.1748 0.9000 0.9023   1.00  1  1  3.0E-05
  9 :  -2.27E-02 8.49E-07 0.062 0.1770 0.9040 0.9000   1.00  1  1  4.5E-06
 10 :  -2.27E-02 1.75E-07 0.179 0.2064 0.9071 0.9000   1.00  1  1  4.5E-07
 11 :  -2.27E-02 3.98E-08 0.113 0.2271 0.9116 0.9000   1.00  1  1  7.1E-08
 12 :  -2.27E-02 8.36E-09 0.000 0.2101 0.9054 0.9000   1.00  1  1  1.5E-08
 13 :  -2.27E-02 1.53E-09 0.000 0.1832 0.9000 0.9045   1.00  1  1  8.8E-09

iter seconds digits       c*x               b*y
 13      0.1   7.4 -2.2688341674e-02 -2.2688342628e-02
|Ax-b| =   4.0e-09, [Ay-c]_+ =   2.3E-09, |x|=  1.5e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    7.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0226883
Iteration 15, residual norm 0.0226883
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.21E-02 2.25E-01 0.000 0.6016 0.9000 0.9000   5.30  1  1  1.0E+00
  2 :  -3.13E-01 6.93E-02 0.000 0.3075 0.9000 0.9000   2.08  1  1  2.1E-01
  3 :  -9.19E-02 2.29E-02 0.000 0.3304 0.9000 0.9000   1.25  1  1  6.2E-02
  4 :  -2.86E-02 7.24E-03 0.000 0.3163 0.9000 0.9000   1.12  1  1  1.9E-02
  5 :  -2.13E-02 1.48E-03 0.000 0.2046 0.9000 0.9000   1.06  1  1  7.2E-03
  6 :  -2.06E-02 3.81E-04 0.000 0.2573 0.9000 0.9000   1.03  1  1  3.8E-03
  7 :  -2.06E-02 6.87E-05 0.000 0.1802 0.9000 0.9073   1.01  1  1  1.6E-03
  8 :  -2.06E-02 1.14E-05 0.000 0.1665 0.9000 0.9100   1.01  1  1  4.3E-04
  9 :  -2.06E-02 2.07E-06 0.000 0.1812 0.7581 0.9000   1.00  1  1  9.2E-05
 10 :  -2.06E-02 4.10E-08 0.078 0.0198 0.9900 0.9902   1.00  1  1  2.0E-06
 11 :  -2.06E-02 2.25E-09 0.157 0.0548 0.9000 0.9096   1.00  1  1  4.9E-08
 12 :  -2.06E-02 9.63E-11 0.490 0.0428 0.9900 0.9900   1.00  1  1  2.1E-09

iter seconds digits       c*x               b*y
 12      0.1   Inf -2.0601422430e-02 -2.0601421950e-02
|Ax-b| =   1.3e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    7.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.91594.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0206014
Iteration 16, residual norm 0.0206014
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   6.85E-02 2.16E-01 0.000 0.5777 0.9000 0.9000   5.41  1  1  9.1E-01
  2 :   3.43E-02 6.65E-02 0.000 0.3072 0.9000 0.9000   2.15  1  1  1.8E-01
  3 :   3.74E-03 1.51E-02 0.000 0.2267 0.9000 0.9000   1.44  1  1  3.3E-02
  4 :  -1.30E-02 4.04E-03 0.000 0.2680 0.9000 0.9000   1.17  1  1  8.2E-03
  5 :  -1.74E-02 8.07E-04 0.000 0.1999 0.9000 0.9000   1.06  1  1  1.6E-03
  6 :  -1.84E-02 1.70E-04 0.000 0.2111 0.9000 0.9000   1.02  1  1  3.3E-04
  7 :  -1.86E-02 2.99E-05 0.000 0.1757 0.9000 0.9000   1.01  1  1  6.4E-05
  8 :  -1.87E-02 6.00E-06 0.167 0.2004 0.9001 0.9000   1.00  1  1  2.0E-05
  9 :  -1.87E-02 1.26E-06 0.211 0.2098 0.9031 0.9000   1.00  1  1  3.4E-06
 10 :  -1.87E-02 2.68E-07 0.175 0.2132 0.9056 0.9000   1.00  1  1  4.9E-07
 11 :  -1.87E-02 5.56E-08 0.129 0.2074 0.9045 0.9000   1.00  1  1  1.0E-07
 12 :  -1.87E-02 1.10E-08 0.118 0.1978 0.9011 0.9000   1.00  1  1  2.3E-08
 13 :  -1.87E-02 2.19E-09 0.169 0.1992 0.9000 0.9017   1.00  1  1  1.1E-08

iter seconds digits       c*x               b*y
 13      0.1   7.4 -1.8687617411e-02 -1.8687618262e-02
|Ax-b| =   8.0e-09, [Ay-c]_+ =   3.2E-09, |x|=  1.5e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
2.000E-02    7.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0186876
Iteration 17, residual norm 0.0186876
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.18E-02 2.25E-01 0.000 0.6017 0.9000 0.9000   5.30  1  1  1.0E+00
  2 :  -3.11E-01 6.95E-02 0.000 0.3085 0.9000 0.9000   2.08  1  1  2.1E-01
  3 :  -8.79E-02 2.30E-02 0.000 0.3307 0.9000 0.9000   1.25  1  1  6.3E-02
  4 :  -2.50E-02 7.23E-03 0.000 0.3144 0.9000 0.9000   1.12  1  1  1.9E-02
  5 :  -1.71E-02 1.49E-03 0.000 0.2059 0.9000 0.9000   1.06  1  1  7.5E-03
  6 :  -1.70E-02 3.82E-04 0.000 0.2566 0.9000 0.9000   1.03  1  1  4.2E-03
  7 :  -1.70E-02 7.69E-05 0.000 0.2013 0.4239 0.9000   1.01  1  1  2.4E-03
  8 :  -1.71E-02 1.25E-05 0.000 0.1625 0.9000 0.9092   1.01  1  1  6.5E-04
  9 :  -1.70E-02 3.61E-07 0.000 0.0289 0.9900 0.9901   1.00  1  1  1.9E-05
 10 :  -1.70E-02 5.41E-09 0.385 0.0150 0.9900 0.9901   1.00  1  1  1.3E-07
 11 :  -1.70E-02 2.99E-10 0.000 0.0553 0.8958 0.9900   1.00  1  1  7.1E-09

iter seconds digits       c*x               b*y
 11      0.1   Inf -1.6964016229e-02 -1.6964013798e-02
|Ax-b| =   4.4e-08, [Ay-c]_+ =   9.2E-10, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    6.000E-02    1.000E-02    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.9112.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.016964
Iteration 18, residual norm 0.016964
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   6.94E-02 2.17E-01 0.000 0.5784 0.9000 0.9000   5.41  1  1  9.1E-01
  2 :   3.59E-02 6.69E-02 0.000 0.3090 0.9000 0.9000   2.15  1  1  1.8E-01
  3 :   6.66E-03 1.50E-02 0.000 0.2244 0.9000 0.9000   1.44  1  1  3.3E-02
  4 :  -9.85E-03 3.76E-03 0.000 0.2507 0.9000 0.9000   1.17  1  1  7.6E-03
  5 :  -1.41E-02 7.66E-04 0.000 0.2034 0.9000 0.9000   1.05  1  1  1.5E-03
  6 :  -1.51E-02 1.60E-04 0.000 0.2084 0.9000 0.9000   1.02  1  1  3.1E-04
  7 :  -1.53E-02 3.09E-05 0.000 0.1939 0.9023 0.9000   1.01  1  1  5.9E-05
  8 :  -1.54E-02 6.32E-06 0.078 0.2044 0.9022 0.9000   1.00  1  1  1.3E-05
  9 :  -1.54E-02 1.25E-06 0.092 0.1981 0.9017 0.9000   1.00  1  1  4.7E-06
 10 :  -1.54E-02 2.43E-07 0.131 0.1944 0.9002 0.9000   1.00  1  1  1.4E-06
 11 :  -1.54E-02 4.83E-08 0.194 0.1986 0.9000 0.9009   1.00  1  1  4.0E-07
 12 :  -1.54E-02 9.99E-09 0.252 0.2067 0.9000 0.9018   1.00  1  1  1.1E-07
 13 :  -1.54E-02 2.06E-09 0.250 0.2064 0.9000 0.9022   1.00  1  1  2.8E-08
 14 :  -1.54E-02 4.19E-10 0.240 0.2033 0.9000 0.9015   1.00  1  1  6.4E-09

iter seconds digits       c*x               b*y
 14      0.1   7.4 -1.5392805700e-02 -1.5392806284e-02
|Ax-b| =   2.1e-09, [Ay-c]_+ =   4.7E-10, |x|=  1.6e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    8.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0153928
Iteration 19, residual norm 0.0153928
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.15E-02 2.25E-01 0.000 0.6018 0.9000 0.9000   5.31  1  1  1.0E+00
  2 :  -3.08E-01 6.97E-02 0.000 0.3093 0.9000 0.9000   2.08  1  1  2.1E-01
  3 :  -8.49E-02 2.31E-02 0.000 0.3310 0.9000 0.9000   1.25  1  1  6.3E-02
  4 :  -2.22E-02 7.21E-03 0.000 0.3126 0.9000 0.9000   1.12  1  1  1.9E-02
  5 :  -1.37E-02 1.47E-03 0.000 0.2035 0.9000 0.9000   1.06  1  1  8.7E-03
  6 :  -1.41E-02 3.68E-04 0.000 0.2510 0.9000 0.9000   1.03  1  1  5.0E-03
  7 :  -1.40E-02 6.77E-05 0.000 0.1839 0.4806 0.9000   1.01  1  1  2.3E-03
  8 :  -1.40E-02 3.03E-06 0.000 0.0447 0.9900 0.9905   1.00  1  1  2.4E-04
  9 :  -1.40E-02 3.85E-08 0.000 0.0127 0.9537 0.9900   1.00  1  1  9.4E-07
 10 :  -1.40E-02 6.48E-09 0.050 0.1683 0.9000 0.9017   1.00  1  1  1.6E-07
 11 :  -1.40E-02 1.06E-09 0.069 0.1629 0.8872 0.9000   1.00  1  1  2.6E-08
 12 :  -1.40E-02 2.19E-10 0.188 0.2078 0.9000 0.9000   1.00  1  1  5.4E-09

iter seconds digits       c*x               b*y
 12      0.1   Inf -1.3979503180e-02 -1.3979502226e-02
|Ax-b| =   3.4e-08, [Ay-c]_+ =   2.9E-10, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    7.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.90793.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0139795
Iteration 20, residual norm 0.0139795
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.02E-02 2.17E-01 0.000 0.5790 0.9000 0.9000   5.42  1  1  9.1E-01
  2 :   3.72E-02 6.73E-02 0.000 0.3105 0.9000 0.9000   2.15  1  1  1.8E-01
  3 :   8.60E-03 1.49E-02 0.000 0.2210 0.9000 0.9000   1.44  1  1  3.2E-02
  4 :  -7.67E-03 3.13E-03 0.000 0.2107 0.9000 0.9000   1.17  1  1  6.3E-03
  5 :  -1.15E-02 6.58E-04 0.000 0.2099 0.9000 0.9000   1.04  1  1  1.3E-03
  6 :  -1.25E-02 1.27E-04 0.000 0.1936 0.9063 0.9000   1.02  1  1  2.3E-04
  7 :  -1.26E-02 2.97E-05 0.042 0.2328 0.9083 0.9000   1.00  1  1  4.9E-05
  8 :  -1.27E-02 7.59E-06 0.000 0.2558 0.9107 0.9000   1.00  1  1  1.2E-05
  9 :  -1.27E-02 1.39E-06 0.000 0.1833 0.9000 0.9042   1.00  1  1  1.2E-05
 10 :  -1.27E-02 1.99E-07 0.000 0.1432 0.9000 0.5298   1.00  1  1  8.7E-07
 11 :  -1.27E-02 3.52E-08 0.000 0.1766 0.9000 0.9000   1.00  1  1  1.5E-07
 12 :  -1.27E-02 7.90E-09 0.029 0.2247 0.9000 0.9000   1.00  1  1  3.4E-08
 13 :  -1.27E-02 1.41E-09 0.000 0.1781 0.9000 0.9000   1.00  1  1  6.1E-09

iter seconds digits       c*x               b*y
 13      0.1   Inf -1.2697803722e-02 -1.2697799427e-02
|Ax-b| =   3.7e-09, [Ay-c]_+ =   6.5E-09, |x|=  1.6e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    8.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0126978
Iteration 21, residual norm 0.0126978
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.13E-02 2.25E-01 0.000 0.6019 0.9000 0.9000   5.31  1  1  1.0E+00
  2 :  -3.07E-01 6.99E-02 0.000 0.3099 0.9000 0.9000   2.08  1  1  2.1E-01
  3 :  -8.26E-02 2.31E-02 0.000 0.3313 0.9000 0.9000   1.25  1  1  6.3E-02
  4 :  -2.00E-02 7.20E-03 0.000 0.3110 0.9000 0.9000   1.12  1  1  1.9E-02
  5 :  -1.11E-02 1.52E-03 0.000 0.2106 0.9000 0.9000   1.06  1  1  7.6E-03
  6 :  -1.17E-02 3.64E-04 0.000 0.2399 0.9000 0.9000   1.03  1  1  5.5E-03
  7 :  -1.16E-02 5.63E-05 0.000 0.1548 0.4533 0.9000   1.01  1  1  2.1E-03
  8 :  -1.16E-02 1.94E-06 0.000 0.0344 0.9900 0.9905   1.00  1  1  1.8E-04
  9 :  -1.15E-02 3.09E-08 0.141 0.0160 0.9585 0.9900   1.00  1  1  7.8E-07
 10 :  -1.15E-02 3.47E-09 0.242 0.1124 0.9450 0.9474   1.00  1  1  8.8E-08
 11 :  -1.15E-02 2.88E-10 0.184 0.0828 0.9465 0.9450   1.00  1  1  7.3E-09

iter seconds digits       c*x               b*y
 11      0.1   Inf -1.1544677812e-02 -1.1544674812e-02
|Ax-b| =   4.9e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    6.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.9048.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0115447
Iteration 22, residual norm 0.0115447
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.09E-02 2.17E-01 0.000 0.5794 0.9000 0.9000   5.42  1  1  9.1E-01
  2 :   3.83E-02 6.77E-02 0.000 0.3118 0.9000 0.9000   2.15  1  1  1.8E-01
  3 :   9.90E-03 1.47E-02 0.000 0.2178 0.9000 0.9000   1.45  1  1  3.2E-02
  4 :  -5.22E-03 3.09E-03 0.000 0.2097 0.9000 0.9000   1.17  1  1  6.2E-03
  5 :  -9.07E-03 8.27E-04 0.000 0.2674 0.9000 0.9000   1.03  1  1  1.6E-03
  6 :  -1.04E-02 7.51E-05 0.442 0.0909 0.9740 0.9900   1.02  1  1  1.3E-04
  7 :  -1.05E-02 1.81E-05 0.119 0.2410 0.9000 0.9019   1.00  1  1  5.4E-05
  8 :  -1.05E-02 3.49E-06 0.000 0.1928 0.9000 0.9048   1.00  1  1  5.6E-05
  9 :  -1.05E-02 6.13E-07 0.081 0.1757 0.9000 0.9042   1.00  1  1  1.5E-05
 10 :  -1.05E-02 1.17E-07 0.250 0.1915 0.9000 0.9044   1.00  1  1  3.5E-06
 11 :  -1.05E-02 2.22E-08 0.246 0.1894 0.9000 0.9050   1.00  1  1  7.9E-07
 12 :  -1.05E-02 4.00E-09 0.231 0.1797 0.9000 0.9021   1.00  1  1  1.5E-07
 13 :  -1.05E-02 7.03E-10 0.240 0.1760 0.9007 0.9000   1.00  1  1  2.6E-08
 14 :  -1.05E-02 1.23E-10 0.244 0.1742 0.9029 0.9000   1.00  1  1  4.3E-09

iter seconds digits       c*x               b*y
 14      0.1   7.6 -1.0501636113e-02 -1.0501636399e-02
|Ax-b| =   6.5e-10, [Ay-c]_+ =   8.2E-11, |x|=  1.6e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    9.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0105016
Iteration 23, residual norm 0.0105016
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.11E-02 2.25E-01 0.000 0.6020 0.9000 0.9000   5.31  1  1  1.0E+00
  2 :  -3.05E-01 7.00E-02 0.000 0.3104 0.9000 0.9000   2.08  1  1  2.1E-01
  3 :  -8.08E-02 2.32E-02 0.000 0.3315 0.9000 0.9000   1.25  1  1  6.3E-02
  4 :  -1.83E-02 7.18E-03 0.000 0.3096 0.9000 0.9000   1.12  1  1  1.9E-02
  5 :  -9.04E-03 1.60E-03 0.000 0.2231 0.9000 0.9000   1.06  1  1  4.1E-03
  6 :  -9.78E-03 3.64E-04 0.000 0.2271 0.9000 0.9000   1.03  1  1  6.0E-03
  7 :  -9.59E-03 2.57E-05 0.217 0.0707 0.6754 0.9900   1.01  1  1  1.1E-03
  8 :  -9.59E-03 3.64E-06 0.000 0.1415 0.9000 0.9199   1.00  1  1  3.0E-04
  9 :  -9.56E-03 1.39E-07 0.463 0.0382 0.9900 0.9901   1.00  1  1  5.8E-06
 10 :  -9.56E-03 2.76E-08 0.158 0.1983 0.7014 0.9000   1.00  1  1  7.1E-07
 11 :  -9.56E-03 4.63E-09 0.021 0.1678 0.9000 0.9035   1.00  1  1  1.2E-07
 12 :  -9.56E-03 8.96E-10 0.217 0.1937 0.9000 0.9058   1.00  1  1  2.3E-08
 13 :  -9.56E-03 2.04E-10 0.330 0.2280 0.9000 0.9107   1.00  1  1  5.3E-09

iter seconds digits       c*x               b*y
 13      0.1   Inf -9.5619409413e-03 -9.5619353033e-03
|Ax-b| =   3.7e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    9.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.90335.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.00956194
Iteration 24, residual norm 0.00956194
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.15E-02 2.17E-01 0.000 0.5798 0.9000 0.9000   5.42  1  1  9.1E-01
  2 :   3.92E-02 6.79E-02 0.000 0.3130 0.9000 0.9000   2.15  1  1  1.8E-01
  3 :   1.08E-02 1.46E-02 0.000 0.2153 0.9000 0.9000   1.45  1  1  3.2E-02
  4 :  -3.30E-03 3.01E-03 0.000 0.2060 0.9000 0.9000   1.17  1  1  6.0E-03
  5 :  -7.00E-03 9.47E-04 0.000 0.3141 0.9000 0.9000   1.03  1  1  1.9E-03
  6 :  -8.58E-03 6.95E-05 0.325 0.0734 0.9794 0.9900   1.02  1  1  1.2E-04
  7 :  -8.68E-03 1.77E-05 0.128 0.2552 0.9034 0.9000   1.01  1  1  3.0E-05
  8 :  -8.70E-03 3.96E-06 0.000 0.2233 0.9016 0.9000   1.00  1  1  1.9E-05
  9 :  -8.71E-03 7.06E-07 0.000 0.1783 0.9000 0.9032   1.00  1  1  1.2E-05
 10 :  -8.71E-03 1.26E-07 0.101 0.1779 0.9000 0.9023   1.00  1  1  2.9E-06
 11 :  -8.71E-03 2.49E-08 0.257 0.1985 0.9000 0.9041   1.00  1  1  7.3E-07
 12 :  -8.71E-03 4.95E-09 0.253 0.1986 0.9000 0.9050   1.00  1  1  1.8E-07
 13 :  -8.71E-03 9.19E-10 0.238 0.1856 0.9000 0.9025   1.00  1  1  3.5E-08
 14 :  -8.71E-03 1.64E-10 0.241 0.1779 0.9002 0.9000   1.00  1  1  6.2E-09

iter seconds digits       c*x               b*y
 14      0.1   7.4 -8.7126081972e-03 -8.7126085458e-03
|Ax-b| =   5.4e-10, [Ay-c]_+ =   1.0E-10, |x|=  1.6e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    8.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.00871261
Iteration 25, residual norm 0.00871261
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.09E-02 2.25E-01 0.000 0.6021 0.9000 0.9000   5.31  1  1  1.0E+00
  2 :  -3.04E-01 7.01E-02 0.000 0.3108 0.9000 0.9000   2.08  1  1  2.1E-01
  3 :  -7.95E-02 2.32E-02 0.000 0.3318 0.9000 0.9000   1.25  1  1  6.3E-02
  4 :  -1.71E-02 7.17E-03 0.000 0.3084 0.9000 0.9000   1.13  1  1  1.9E-02
  5 :  -7.41E-03 1.67E-03 0.000 0.2325 0.9000 0.9000   1.06  1  1  4.2E-03
  6 :  -8.22E-03 3.57E-04 0.000 0.2145 0.9000 0.9000   1.04  1  1  6.9E-03
  7 :  -7.98E-03 2.09E-05 0.226 0.0586 0.6224 0.9900   1.01  1  1  7.7E-04
  8 :  -7.98E-03 2.84E-06 0.000 0.1355 0.9000 0.9186   1.00  1  1  1.9E-04
  9 :  -7.95E-03 1.06E-07 0.432 0.0372 0.9900 0.9901   1.00  1  1  2.3E-06
 10 :  -7.95E-03 2.41E-08 0.139 0.2281 0.7546 0.9000   1.00  1  1  6.2E-07
 11 :  -7.95E-03 4.17E-09 0.016 0.1730 0.9000 0.9020   1.00  1  1  1.1E-07
 12 :  -7.95E-03 8.10E-10 0.184 0.1943 0.9000 0.9054   1.00  1  1  2.1E-08
 13 :  -7.95E-03 1.82E-10 0.315 0.2253 0.9000 0.9085   1.00  1  1  4.7E-09

iter seconds digits       c*x               b*y
 13      0.1   Inf -7.9462769764e-03 -7.9462722643e-03
|Ax-b| =   3.3e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    8.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.90186.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.00794627
Iteration 26, residual norm 0.00794627
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.21E-02 2.17E-01 0.000 0.5801 0.9000 0.9000   5.42  1  1  9.1E-01
  2 :   3.99E-02 6.82E-02 0.000 0.3139 0.9000 0.9000   2.15  1  1  1.8E-01
  3 :   1.14E-02 1.46E-02 0.000 0.2136 0.9000 0.9000   1.45  1  1  3.1E-02
  4 :  -1.89E-03 2.91E-03 0.000 0.1997 0.9000 0.9000   1.17  1  1  5.8E-03
  5 :  -5.35E-03 9.97E-04 0.000 0.3427 0.9000 0.9000   1.02  1  1  2.0E-03
  6 :  -7.14E-03 6.33E-05 0.093 0.0635 0.9735 0.9900   1.02  1  1  1.1E-04
  7 :  -7.22E-03 1.60E-05 0.000 0.2527 0.9007 0.9000   1.01  1  1  9.2E-05
  8 :  -7.25E-03 3.55E-06 0.000 0.2218 0.9000 0.9003   1.00  1  1  4.7E-05
  9 :  -7.25E-03 6.78E-07 0.007 0.1910 0.9000 0.9012   1.00  1  1  1.3E-05
 10 :  -7.25E-03 1.28E-07 0.169 0.1882 0.9000 0.9010   1.00  1  1  3.0E-06
 11 :  -7.25E-03 2.45E-08 0.226 0.1916 0.9000 0.9006   1.00  1  1  6.3E-07
 12 :  -7.25E-03 4.61E-09 0.247 0.1884 0.9008 0.9000   1.00  1  1  1.2E-07
 13 :  -7.25E-03 8.44E-10 0.246 0.1831 0.9021 0.9000   1.00  1  1  1.9E-08
 14 :  -7.25E-03 1.53E-10 0.231 0.1811 0.9023 0.9000   1.00  1  1  2.8E-09

iter seconds digits       c*x               b*y
 14      0.1   7.8 -7.2533995096e-03 -7.2533996309e-03
|Ax-b| =   4.8e-10, [Ay-c]_+ =   2.5E-10, |x|=  1.6e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    1.000E-01    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.0072534
Iteration 27, residual norm 0.0072534
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.08E-02 2.25E-01 0.000 0.6022 0.9000 0.9000   5.31  1  1  1.0E+00
  2 :  -3.03E-01 7.01E-02 0.000 0.3110 0.9000 0.9000   2.08  1  1  2.1E-01
  3 :  -7.85E-02 2.33E-02 0.000 0.3320 0.9000 0.9000   1.25  1  1  6.3E-02
  4 :  -1.61E-02 7.16E-03 0.000 0.3074 0.9000 0.9000   1.13  1  1  1.9E-02
  5 :  -6.12E-03 1.71E-03 0.000 0.2392 0.9000 0.9000   1.06  1  1  4.3E-03
  6 :  -6.95E-03 3.43E-04 0.000 0.2004 0.9000 0.9000   1.04  1  1  8.3E-03
  7 :  -6.66E-03 1.76E-05 0.247 0.0513 0.5586 0.9900   1.01  1  1  3.7E-04
  8 :  -6.66E-03 2.31E-06 0.000 0.1317 0.9000 0.9171   1.00  1  1  6.4E-05
  9 :  -6.63E-03 1.59E-07 0.411 0.0688 0.9900 0.9900   1.00  1  1  3.1E-06
 10 :  -6.63E-03 3.43E-08 0.159 0.2155 0.7151 0.9000   1.00  1  1  8.9E-07
 11 :  -6.63E-03 3.69E-09 0.183 0.1073 0.9450 0.9463   1.00  1  1  9.6E-08
 12 :  -6.63E-03 7.07E-10 0.202 0.1918 0.9000 0.9077   1.00  1  1  1.8E-08
 13 :  -6.63E-03 1.64E-10 0.349 0.2314 0.9000 0.9118   1.00  1  1  4.2E-09

iter seconds digits       c*x               b*y
 13      0.1   Inf -6.6271446296e-03 -6.6271383826e-03
|Ax-b| =   2.9e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
0.000E+00    8.000E-02    0.000E+00    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.90077.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.00662714
Iteration 28, residual norm 0.00662714
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.26E-02 2.17E-01 0.000 0.5803 0.9000 0.9000   5.42  1  1  9.1E-01
  2 :   4.05E-02 6.84E-02 0.000 0.3147 0.9000 0.9000   2.15  1  1  1.8E-01
  3 :   1.18E-02 1.45E-02 0.000 0.2125 0.9000 0.9000   1.45  1  1  3.1E-02
  4 :  -8.81E-04 2.79E-03 0.000 0.1916 0.9000 0.9000   1.18  1  1  5.5E-03
  5 :  -4.12E-03 9.70E-04 0.000 0.3483 0.9000 0.9000   1.02  1  1  1.9E-03
  6 :  -5.97E-03 5.50E-05 0.000 0.0567 0.9801 0.9900   1.02  1  1  9.0E-05
  7 :  -6.04E-03 1.47E-05 0.004 0.2676 0.9000 0.9057   1.01  1  1  2.4E-04
  8 :  -6.06E-03 3.37E-06 0.000 0.2288 0.9000 0.9037   1.00  1  1  1.0E-04
  9 :  -6.06E-03 5.84E-07 0.000 0.1733 0.9000 0.9001   1.00  1  1  2.0E-05
 10 :  -6.06E-03 9.07E-08 0.000 0.1552 0.9022 0.9000   1.00  1  1  2.9E-06
 11 :  -6.06E-03 1.37E-08 0.113 0.1516 0.9052 0.9000   1.00  1  1  2.7E-07
 12 :  -6.06E-03 1.99E-09 0.165 0.1447 0.9089 0.9000   1.00  1  1  6.0E-09

iter seconds digits       c*x               b*y
 12      0.1   Inf -6.0604095495e-03 -6.0604083794e-03
|Ax-b| =   6.7e-09, [Ay-c]_+ =   5.8E-09, |x|=  1.6e+00, |y|=  3.6e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    6.000E-02    1.000E-02    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.00606041
Iteration 29, residual norm 0.00606041
 
Calling sedumi: 151 variables, 51 equality constraints
   For improved efficiency, sedumi is solving the dual problem.
------------------------------------------------------------
SeDuMi 1.21 by AdvOL, 2005-2008 and Jos F. Sturm, 1998-2003.
Alg = 2: xz-corrector, Adaptive Step-Differentiation, theta = 0.250, beta = 0.500
eqs m = 51, order n = 53, dim = 152, blocks = 2
nnz(A) = 550 + 1, nnz(ADA) = 250, nnz(L) = 151
Handling 1 + 1 dense columns.
 it :     b*y       gap    delta  rate   t/tP*  t/tD*   feas cg cg  prec
  0 :            3.74E-01 0.000
  1 :   7.07E-02 2.26E-01 0.000 0.6022 0.9000 0.9000   5.31  1  1  1.0E+00
  2 :  -3.03E-01 7.02E-02 0.000 0.3112 0.9000 0.9000   2.08  1  1  2.1E-01
  3 :  -7.78E-02 2.33E-02 0.000 0.3322 0.9000 0.9000   1.25  1  1  6.4E-02
  4 :  -1.54E-02 7.15E-03 0.000 0.3066 0.9000 0.9000   1.13  1  1  1.9E-02
  5 :  -5.10E-03 1.74E-03 0.000 0.2439 0.9000 0.9000   1.06  1  1  4.4E-03
  6 :  -5.91E-03 3.17E-04 0.000 0.1819 0.9000 0.9000   1.04  1  1  1.0E-02
  7 :  -5.58E-03 1.51E-05 0.270 0.0475 0.4888 0.9900   1.01  1  1  2.2E-04
  8 :  -5.58E-03 1.95E-06 0.000 0.1293 0.9000 0.9156   1.00  1  1  4.4E-05
  9 :  -5.55E-03 3.03E-07 0.000 0.1555 0.9000 0.9093   1.00  1  1  8.0E-06
 10 :  -5.55E-03 3.48E-08 0.325 0.1150 0.9450 0.9459   1.00  1  1  9.2E-07
 11 :  -5.55E-03 7.40E-09 0.311 0.2125 0.9000 0.9088   1.00  1  1  1.9E-07
 12 :  -5.55E-03 1.47E-09 0.269 0.1980 0.9000 0.9068   1.00  1  1  3.8E-08
 13 :  -5.55E-03 2.61E-10 0.000 0.1778 0.9000 0.6256   1.00  1  1  7.0E-09

iter seconds digits       c*x               b*y
 13      0.1   Inf -5.5474670964e-03 -5.5474655447e-03
|Ax-b| =   4.8e-08, [Ay-c]_+ =   0.0E+00, |x|=  1.4e+00, |y|=  4.9e+00

Detailed timing (sec)
   Pre          IPM          Post
1.000E-02    7.000E-02    1.000E-02    
Max-norms: ||b||=1, ||c|| = 2.360448e+00,
Cholesky |add|=0, |skip| = 1, ||L.L|| = 1.90009.
------------------------------------------------------------
Status: Solved
Optimal value (cvx_optval): +0.00554747
Iteration 30, residual norm 0.00554747
Original matrix:
    0.3592    0.8312    0.8065    0.9996    0.5081    1.3236    0.8398    1.2189    0.6605    1.0759
    0.4201    0.6033    0.5158    0.9656    0.6029    1.1994    1.0734    1.2186    0.3909    0.7756
    0.4017    1.3395    0.8938    1.6807    1.0275    2.1509    1.5564    2.0325    0.9134    1.8006
    0.4893    1.0137    0.7657    1.4035    0.7798    1.5208    1.3870    2.0219    0.7193    1.4301
    0.4655    0.6601    0.4999    1.2935    0.8624    1.5954    1.4953    1.3942    0.3916    0.8624
    0.2848    0.7062    0.3622    1.2739    0.8833    1.5680    1.4027    1.2944    0.4016    0.9674
    0.5641    1.2552    0.8794    1.9054    1.2033    2.3604    1.9676    2.2161    0.8212    1.6840
    0.4592    0.8008    0.8972    1.1923    0.6308    1.6067    1.0947    1.2161    0.6544    1.0043
    0.4199    0.8984    0.6402    1.1014    0.7246    1.5146    1.1439    1.4494    0.5609    1.1297
    0.6042    0.7117    0.9137    1.4382    0.7149    1.5961    1.4474    1.5962    0.6271    0.9826

Left factor Y:
    0.4286    1.1793    0.0000    0.6660    0.0137
    0.6350    0.0788    0.4830    0.4518    0.5320
    0.0171    1.2370    0.3798    1.0579    0.5409
    0.5301    0.5462    1.1912    1.8075    0.2256
    0.5708    0.0000    0.7478    0.0251    0.8437
    0.0000    0.2052    0.6874    0.0101    0.8187
    0.4129    0.7281    0.8462    0.7716    0.8623
    0.6630    1.2418    0.0000    0.0634    0.1579
    0.4666    0.3116    0.0544    0.6028    0.7242
    1.0270    0.9299    1.1089    0.6148    0.0000

Right factor X:
    0.3474    0.0576    0.3912    0.2032    0.0403    0.2932    0.2695    0.3074    0.1095   -0.0000
    0.1504    0.5049    0.4657    0.7229    0.3881    0.9549    0.5974    0.6500    0.4118    0.6707
    0.0730   -0.0000    0.0000    0.4630    0.2634    0.3237    0.5510    0.3551    0.0156    0.0965
    0.0441    0.3000    0.1276    0.0705    0.0339    0.0783    0.0073    0.4588    0.1847    0.4093
    0.2496    0.7340    0.3252    0.9848    0.7597    1.4029    1.1011    1.1151    0.3717    0.9252

Residual A - Y * X:
    0.0002    0.0012    0.0001   -0.0005    0.0001    0.0004   -0.0002   -0.0003   -0.0002   -0.0004
   -0.0003    0.0009   -0.0000    0.0003    0.0000   -0.0002   -0.0000    0.0002    0.0002   -0.0009
    0.0003   -0.0005    0.0001   -0.0001   -0.0001   -0.0001    0.0002   -0.0003   -0.0003    0.0008
    0.0000   -0.0003   -0.0000   -0.0001    0.0000    0.0001   -0.0001    0.0000    0.0000    0.0002
    0.0008    0.0004   -0.0010   -0.0014    0.0006    0.0003    0.0002    0.0009   -0.0009   -0.0006
   -0.0011   -0.0014   -0.0009    0.0003    0.0002    0.0002   -0.0003   -0.0006    0.0002    0.0019
    0.0001   -0.0005   -0.0000   -0.0001    0.0000    0.0001   -0.0000   -0.0000   -0.0000    0.0004
   -0.0001    0.0008    0.0000   -0.0001    0.0000    0.0001   -0.0001   -0.0000    0.0000   -0.0006
   -0.0004    0.0018    0.0001    0.0005   -0.0001   -0.0005    0.0002   -0.0000    0.0000   -0.0013
   -0.0005   -0.0013    0.0004    0.0006   -0.0003   -0.0001   -0.0004    0.0002    0.0008    0.0002

Residual after 30 iterations: 0.00554747