decisiontree-ka4ki

st.py

#!/usr/bin/env python2

import numpy as np
from sklearn import tree
from sklearn.tree import DecisionTreeClassifier

X = []
y = []

for line in open("mammographic_masses.data"):
	line = line.split(',')
	if not '?' in line:
		X.append(map(int,line))

X = np.array(X)

print X


XX = X[:,range(0,5)]
YY = X[:,5]

for size in range(6,10):
 
	Xlearn = XX[:len(XX)/10*size]
	Ylearn = YY[:len(YY)/10*size]

	Xtest = XX[len(XX)/10*size:]
	Ytest = YY[len(YY)/10*size:]

	clf = DecisionTreeClassifier().fit(Xlearn, Ylearn)	

	filename = "dtree" + str(size) + ".dot"

	with open(filename, 'w') as f:
		f = tree.export_graphviz(clf, out_file=f)

	Ypredicted = clf.predict(Xtest)	

	rP = 0
	wP = 0

	for i in range(len(XX)/10*(10-size)):
		if Ypredicted[i] == Ytest[i]:
			rP += 1
		else:
			wP += 1

	print "TEST SIZE: ", size*10, "%"
	print "right: ", rP
	print "wrong: ", wP
	print "accuracy: ", float(rP)/(rP+wP)