An efficient MCMC algorithm for graphical model inference using Cycle Basis prior.
>>> from utils.cbmcmc import cbmcmc
>>> import numpy as np
>>> mu, Sigma = (np.zeros(10), np.eye(10))
>>> data = np.random.multivariate_normal(mu, Sigma, 1000)
>>> sampler = cbmcmc(data, it=200, basis='cycle', treeprior='all')
>>> res = sampler.res['SAMPLES']