Table of Contents

Class: TransductiveTrainer Bio/Tools/Classification/SVM.py

Transductive Support Vector Machine Trainer

This is used to train transductive support vector machines described in: Joachims, Thorsten. Transductive Inference for Text Classification using Support Vector Machines. XXX url?

The algorithm is modified from Platt's Sequential Minimal Optimization.

Methods: train Train a new SVM.

Methods   
_C
_E
_examine_example
_is_bound
_take_step
train
  _C  
_C ( self,  i )
  _E  
_E ( self,  i )
  _examine_example  
_examine_example ( self,  i2 )
  _is_bound  
_is_bound (
        self,
        alpha,
        C,
        )
  _take_step  
_take_step (
        self,
        i1,
        i2,
        )
  train  
train (
        self,
        training_set,
        results,
        num_test_cases,
        kernel_fn,
        C,
        C_negtest,
        C_postest,
        epsilon,
        update_fn=None,
        )

train(self, training_set, results, num_test_cases, kernel_fn, C, C_negtest, C_postest, epsilon, update_fn=None) -> SVM

Table of Contents

This document was automatically generated on Mon Jul 1 12:03:03 2002 by HappyDoc version 2.0.1