Udacity - Retrieve Accuracy using Decision Tree

sklearn decision tree
split을 2, 50로 설정하고 accuracy를 비교한다.

import sys from class_vis import prettyPicture from prep_terrain_data import makeTerrainData   import matplotlib.pyplot as plt import numpy as np import pylab as pl   features_train, labels_train, features_test, labels_test = makeTerrainData()   ########################## DECISION TREE ################################# ### your code goes here--now create 2 decision tree classifiers, ### one with min_samples_split=2 and one with min_samples_split=50 ### compute the accuracies on the testing data and store ### the accuracy numbers to acc_min_samples_split_2 and ### acc_min_samples_split_50, respectively from sklearn import tree   clf_2 = tree.DecisionTreeClassifier(min_samples_split=2) clf_2.fit(features_train,labels_train)   acc_min_samples_split_2 = clf_2.score(features_test,labels_test)   clf_50 = tree.DecisionTreeClassifier(min_samples_split=50) clf_50.fit(features_train,labels_train)   acc_min_samples_split_50 = clf_50.score(features_test,labels_test)   def submitAccuracies():   return {"acc_min_samples_split_2":round(acc_min_samples_split_2,3),           "acc_min_samples_split_50":round(acc_min_samples_split_50,3)}   print submitAccuracies() Colored by Color Scripter

cs