These are the fitting and initialization functions used by asggm. These should generally not be used directly.
rCSL(x, iterations = 500, init = NULL, epsilon = 1e-05, ansL = NULL)
genL(kNodes, spP)
genData(L, nSamples)design matrix
number of iterations of the algorithm to run.
optional initialization, for instance, the cholesky of x. If NULL, it defaults to the cholesky of x.
amount to add for numerical stability.
L created by genL
number of samples.
rCSL returns a list with the following components:
rCSL calls the C++ code to compute the Wong EM algorithm. genL and genData are used together to create example data.
Wong, Eleanor, Suyash Awate, and P. Thomas Fletcher. “Adaptive Sparsity in Gaussian Graphical Models.”In Proceedings of the 30th International Conference on Machine Learning (ICML-13), pp. 311-319. 2013.
asggm, which should be used directly instead of these methods