AdaBoost and Decision Stumps

To run, pass an ARFF dataset and a number of boosting rounds. For example:

$ python3 adaboost.py --file weather.nominal.arff --iters 50

Note that, when run directly, adaboost.py is hardcoded to use decision stumps as the underlying "weak" classifier learning algorithm.

adaboost.py implements the AdaBoost algorithm from Freund and Schapire.

decision_stump.py implements decision stumps, a simple classifier learning algorithm based on decision trees.

parse_arff.py is used for parsing ARFF files.

coleerickson/adaboost