tiepvupsu / fista Goto Github PK

in norm1.m I have to replace res = full(sum(abs(vec(X)))); to res = full(sum(abs(X(:))));
and it works!
thanks for sharing

Hi,

Thanks a lot for your code it has really helping us for our research. I tried to add a new regularization term, the total variation, but I got strange result, do you think that it is possible to use this kind of regularization with your implementation?

Thanks a lot !

1. In the article you refer pL = argmin ( ... , that is the minimization problem. But instead of minimizing you use some sort of projection. Why did you make such a replacement?
2. In the article you refer pL = (1.5 1.4), for the LASSO problem. Why you didn't use it?

Hello, how can this method be implemented in the nonnegative lasso problem?

Where is the main program？how do I can run this program?

Can i solve the following problem using FISTA?

minimise over x -> norm((y-Ax)./sqrt(Ax), 2) + lambda(norm(x,2))

Hi @tiepvupsu ,

I am now reproducing the FISTA code based on the paper. I find out that the matrix A defined in the paper is the same as the matrix D in your code. My question is, if I am doing a image restoration task, I need to construct a matrix A as the paper section 5.1 states (A=RW), but in your code, this step is ignored and replaced as D=normc(rand(d,k)), would you please explain how to use your code to reproduce the result shown in the paper? I have tried your way to define D, but thing's not work.

Thank you.

The instructions for building the SPAMS library (for use in windows) references a compile.m file that I cannot find. Can you please clarify the procedure for getting this working on windows?

When I try to run any of the demo_XXX.m files I get the following error 👍

Attempt to execute SCRIPT mexLasso as a function

Do I need to do something to compile the mex functions before running any of the demo code?