Hi, Mattia again.

I feel like I’m a broken record looking for help for these networks but I keep getting struck.

I’m trying to run the non negative transmembrane potentials network using the latest solve_messnarz_ADMM.m i found online here:

“https://github.com/jcollfont/inverseecg/blob/master/solve_messnarz_ADMM.m”

(the script inverse_messnarz_ADMM.m that came within the SCIRunDATA archive was outdated and giving “too many input arguments” error, and it wasn’t even starting).

This is the error I’m getting now:

.

.

.

Solving inverse problem for lambda -0.273

Solving inverse problem for lambda -0.227

Solving inverse problem for lambda -0.182

Solving inverse problem for lambda -0.136

Solving inverse problem for lambda -0.091

Solving inverse problem for lambda -0.045

Solving inverse problem for lambda 0.000

Elapsed time is 445.924574 seconds.

Error using corner>parseInputs (line 192)

The value of 'N' is invalid. Expected N to be a scalar.

Error in corner (line 81)

    filter_coef,max_corners,quality_level] = parseInputs(args{:});

Error in l_corner (line 132)

    reg_c = corner(rho,eta);

Error in scirun_code (line 122)

                                lambda_corner = l_corner(logRes(end:-1:1)',logReg(end:-1:1)',lambda(end:-1:1));

Any suggestion of how to fix this?

Many thanks in advance

Mattia Gentile.