mml_rand_str_adaptive: MML random Markov blanket models using adaptive code approach
In kelvinyangli/mbmml: Learn Markov blankets using Minimum Message Length

This function takes a list of given random Markov blanket models and calculates the mml score of the target under each Markov blanket model. The output is a weighted average over all message lengths.

1
2
3

mml_rand_str_adaptive(data, vars, arities, sampleSize, varCnt, targetIndex,
  logProbTarget, cachedPXGivenT, probsMtx, strList, mbIndices, weights,
  cachedPXGivenY, cachInd, alpha, statingPara, debug = FALSE)

`data`	A dataset whost variables are in numeric/integer format. Any categorical variables must be converted into numeric/integer first.
`vars`	A vector of all variables in data, in the same order as the column names of data.
`arities`	A vector of variable arities in data, in the same order as the column names of data.
`sampleSize`	The sample size. That is, the number of rows of data.
`varCnt`	This parameter is for mml_cpt. As explained by argument name. It is obtained by getting the detailed information of the given data using the function count_occurance().
`targetIndex`	The target node's index in vars, whose Markov blanket we are interested in.
`logProbTarget`	Log of the probability of the target.
`cachedPXGivenT`	cachedPXGivenT
`probsMtx`	probsMtx
`strList`	A list of random (or all) Markov blanket models (structures) to go through. Each structure is stored in a matrix format.
`mbIndices`	The indices of potential Markov blanket variables.
`weights`	A vector of weights for random Markov blanket models.
`cachedPXGivenY`	cachedPXGivenY
`cachInd`	cachInd
`alpha`	A vector of concentration parameters for a Dirichlet distribution. Range is from zeor to positive infinity, length is equal to the arity of the target variable.
`statingPara`	If TRUE MML estimate of the parameters are also stated with extra 0.5log(pi*e/6) per parameter, otherwise 0.
`debug`	A boolean argument to show the detailed Markov blanket inclusion steps based on each mml score.

The function outputs the weighted average message length over all given structures. Noticing it is the probabilies over all models are averaged then use this averaged probability to calculate the averaged message length, not to average the message lengths directly. This is done by the function msg_len_ave() function.

kelvinyangli/mbmml documentation built on June 29, 2020, 3:12 a.m.