data of CSDCG method
Description
Data of the conjugate gradient method in the article titled "Hypergraph-based convex semi-supervised unconstrained symmetric matrix factorization for image clustering". CSDCG(1).csv, CSDCG(gg).csv, CSDCG(-sg).csv, and CSDCG(sy).csv respectively correspond to the cases of $\Im$=1, $\Im$=gg, $\Im$=-sg, and $\Im$=sy of the second method proposed in this paper. ACGSSV.csv, ADHCG.csv, ADHCG.csv, and NACGM.csv are other CG methods for comparison, GD.csv is the Gradient descent. All the data are generated by the algorithm proposed in the paper under CUTEst library. --Problem Indicates the name of the problem in the CUTEst library --Dims represent the dimension of the problem --Method Indicates the method used to solve the problem --x is the optimal solution --Func is the value of the function at x --Grad represents the infinite norm of the gradient of the function at x --Nfunc represents the number of function evaluations --Ngrad represents the number of gradient evaluation --Iter represents the number of iterations --Seconds indicate the running time --StateCode indicates whether the status is abnormal, where 0 indicates success.