get_envrdata: Construction of the data for the co-clustering procedures

In Rcoclust: Co-Clustering with Document-Term Matrix

Description

Function which takes as en entry a matrix A_ijx in a sparse format for the rows, and adds in a variable of type environment: the data vectors in format coordinate list for the columns while aggregating the number of rows (n), columns (d), non null elements in total (nnz), per rows (nnzi) and per columns (nnzj), plus the name and the true labels (lbs) of the corresponding dataset.

Usage

get_envrdata(A_ijx,lbs,name,datacol)

Arguments

A_ijx	an object of class matrix with three columns corresponding to the format called coordinate list (COO) (row, column, value).
lbs	integer vector with the true class labels for the rows.
name	char array with the name of the dataset.
datacol	integer with value 0 or 1 if it is added the three vectors for the column-wise data.

Value

As a result an object of class environment which contains the data for the co-clustering procedures.

Examples

  library(Rcoclust);
  
  #load data
  data(data_cstr);
  
  #library(Matrix); 
  #A = sparseMatrix(i=A_ijx[,1]+1,j=A_ijx[,2]+1,
  #                  x=A_ijx[,3]);
  envrdata = get_envrdata(A_ijx,lbs,name,1);
  
  #retrieve matrix size and number of classes
  n=envrdata$n;
  d=envrdata$d;
  g=length(unique(envrdata$lbs));

Rcoclust documentation built on May 6, 2017, 3:34 a.m.