MBASIC.MADBayes.full: MAD-Bayes method to fit the MBASIC model.

In chandlerzuo/mbasic: Matrix Based Analysis for State-space Inference and Clustering.

Description

MAD-Bayes method to fit the MBASIC model.

Usage

MBASIC.MADBayes.full(Y, Gamma = NULL, fac, lambdaw = NULL, lambda = NULL,
  maxitr = 30, S = 2, tol = 1e-10, ncores = 15, nfits = 1,
  nlambdas = 30, para = NULL, initialize = "kmeans")

Arguments

Y	An N by I matrix containing the data from N experiments across I observation units (loci).
Gamma	An N by I matrix for the prior estimated mean for the background state, for N experiments across the I observation units (loci).
fac	A vector of length N denoting the experimental condition for each replicate.
maxitr	The maximum number of iterations in the E-M algorithm. Default: 100.
S	The number of different states.
tol	Tolerance for error in checking the E-M algorithm's convergence. Default: 1e-04.
ncores	The number of CPUs to be used for parallelization.
nfits	The number of random restarts of the model.
lambdap,lambdaw,lambda	Tuning parameters.
family	The distribution of family to be used. Either "lognormal" or "negbin". See details for more information.

Details

TODO.

Value

A list object including the following fields:

allFits	A list of MBASICFit objects for the best model fit with each lambda.
lambda	A vector of all lambdas corresponding to allFits.
Loss	A vector for the loss corresponding to allFits.
BestFit	The MBASICFit object with largest Silhouette score.
Iter	Number of iterations for BestFit.
Time	Time in seconds used to fit the model.

Author(s)

Chandler Zuo [e-mail]

chandlerzuo/mbasic documentation built on May 13, 2019, 3:24 p.m.