Coefficients-mapping in weka Logistic classifier -

to remove the coefficient used in Weka, there can be Aksplus use Mathos using a coefficient: .

However, all the attributes given in the data set (object object for example) are not used in the classifier.

So my question is - how can I create a mapping of attribute - for example, in the toString () method looks like - coefficient-coefficient?

For example, toString example using I get coefficient following:

I want to be mapping as the following:

{avgP = -9.6225, BysP = 5.3931, degree = 0.0016 ... }

double [] [] examples of back coefficient () is a different set of fractions which You pass objects objects in logistics. I needed to understand how to cope with it. I could not find any way to get information through rational methods, but you can easily duplicate it using the filter and get the same set of properties to work with it.

I am using Weka as a library in JRuby and thus here the code is Ruby syntax.

  Import 'weka.filters.unsupervised.attribute.RemoveUseless' logit_filter = Nikalenuselesknu logit_filter.setInputFormat train_filtered logit_filtered = Filter.useFilter (train_filtered, Logit_filter)   
 The logit_filtered variable is an example compile that is a mirror made by rational but the final shrinkage is to keep the barrier of the logistic in the form of the first element of the diploid [] [] which comes back with coefficients, therefore Let us ignore the first element to properly set the attribute Ahia ...  
  java_array = Logit.coefficients.to_a # Java array Ruby coeffs = java_array.map (& amp; to_a) # convert Ruby Coeffs the second level of the Java array Doing .each_with_index; Arr, index | Next if index == 0 # is blocking "# {logit_filtered.attribute (index-1) .name.to_s}" # {coeffs} "end   
 find this map together Well for me.