Compute the Minimum Spanning Tree (MST) of graph g with weights weights, with constraints such that included_edges should be included while excluded_edges should be excluded.

Learn Chow Liu Tree(s). It will run on CPU/GPU based on where train_x is.

Arguments:

• train_x: training data. If want gpu, move to gpu before calling this CuArray(train_x).

Keyword arguments:

• num_trees=1: number of trees you want to learned.
• dropout_prob=0.0: drop edges with probability dropout_prob when learning maximum spanning tree.
• weights=nothing: weights of samples. Weights are all 1 if nothing.
• pseudocount=0.0: add a total of pseudo count spread out overall all categories.
• Float=Float64: precision. Float32 is faster if train_x a large.

Top k minimum spanning trees for a complete graph with weights as weights http://www.nda.ac.jp/~yamada/paper/enum-mst.pdf