Entropy in decision tree

my_entropy.m

# The main difference between MATLAB bundled entropy function
# and this custom function is that they use a transformation to uint8
# and the bundled entropy() function is used mostly for signal processing
# while I simply use a straightforward solution usefull e.g. for learning trees
 
function f = my_entropy(data, weighted, information_gain)
  # function @f accepts only cell arrays (in argument @data);
  
  # @weighted tells whether return one weighed average entropy per row
  # or return a vector of entropies (one entropy per 1 bucket)
  # moreover, I find vectors as the only representation of 'buckets'
  # in other words, vector = bucket (leaf of decision tree)
  
  # @information_gain tells whether to calculate Kullback–Leibler divergence
  # and treat rows as single states after a transformation, or not
  
  if nargin < 2
    weighted = false;
  end;
  if nargin < 3
    information_gain = false;
  end;
  
  rows = @(x) size(x,1);
  cols = @(x) size(x,2);
  
  if weighted
    weights = [];
  end;
  result = [];
  
  for r = 1:rows(data)
  
    for c = 1:cols(data) # in most cases this will be 1:1

      data{r,c}(data{r,c} == 0) = [];
      omega = sum(data{r,c});
      epsilon = 0;
      
      for b = 1:cols(data{r,c})
        epsilon = epsilon + ( (data{r,c}(b) / omega) * (log2(data{r,c}(b) / omega)) );
      end;
      
      if (-epsilon == 0) entropy = 0; else entropy = -epsilon; end;
      
      if weighted
        result = result + entropy;
      end;
      result = [result entropy];
    
    end;
    
  end;
  
  f = result;
  
end;
 
# test cases
 
cell1 = { [16];[16];[2 2 2 2 2 2 2 2];[12];[16] }
cell2 = { [16],[12];[16],[2];[2 2 2 2 2 2 2 2],[8 8];[12],[8 8];[16],[8 8] }
cell3 = { [16],[3 3];[16],[2];[2 2 2 2 2 2 2 2],[2 2];[12],[2];[16],[2] }
a1 = [ 100 60 ]
a2 = [ 20 60 ]
a3 = [ 80 0 ]
entr = my_entropy({a1;a2;a3}, false)

# end