Add a simple decision tree visualizer

0.9.4

Add simple decision tree visualizer

Fix bug in DecisionTreeBuilder

0.9.3

Fix bug related to missing attribute values

fixed bug that caused failure when only 2 numeric bins are used for numeric splitting

0.9.2

bumped version

added flexibility to the Classifiers class

0.9.1

bumped version

fixed api bug with 0.8.0

In DecisionTreeBuilder, ignoreAttributeProbability(int) should have taken a double as its argument, in 0.8.0. It now does.

Refactored TreeBuilder

The core functionality for building trees has been broken out of the former TreeBuilder into more generic components that can be used to build other kinds of trees apart from decision trees. Unit tests have also been added for these components. Additionally, tree build time appears to have improved by about 15%, and memory usage is decreased.

Mahout / Hadoop are also no longer dependencies, making the jar lighter weight.

Bug fix of 0.7.15

a min function should have been a max

Auto determination of acceptable split imbalance

The parameter imbalancePowerPenalty can be tuned to adjust the penalty a split faces for creating imbalance in the tree.

minSplitFraction added as a hyperparameter to treeBuilder

One can now require a minimum percentage of training data flow through each path in a decision tree split by specifying that fraction in the treeBuilder's setter: minSplitFraction

fixed gain ratio penalty as it should be applied to numeric bins

Merge pull request #123 from sanity/fixNumPenalty

bumped version num