机器学习之使用sklearn实现决策树及原理分析

import pandas

columns = ["age", "workclass", "fnlwgt", "education", "education_num", "marital_status", "occupation", "relationship", "race", "sex", 
           "capital_gain", "capital_loss", "hours_per_week", "native_country", "high_income"]
#income = pandas.read_csv("income.csv", index_col=False,names=columns)
income = pandas.read_csv("D:\\test\machineLearning\income.csv", names=columns)
print(income.head(2))

   age          workclass  fnlwgt   education  education_num  \
0   39          State-gov   77516   Bachelors             13   
1   50   Self-emp-not-inc   83311   Bachelors             13   

        marital_status        occupation    relationship    race    sex  \
0        Never-married      Adm-clerical   Not-in-family   White   Male   
1   Married-civ-spouse   Exec-managerial         Husband   White   Male   

   capital_gain  capital_loss  hours_per_week  native_country high_income  
0          2174             0              40   United-States       <=50K  
1             0             0              13   United-States       <=50K  

columns = ["education","marital_status","occupation","relationship","race", "sex", "native_country", "high_income"]
for name in columns:
    #将不同类别转化为数字
    col = pandas.Categorical.from_array(income[name])
    income[name]=col.codes

import numpy
import math

numpy.random.seed(10)
#洗牌
income=income.reindex(numpy.random.permutation(income.index))
#求最接近它的一个整数
train_max_row = math.floor(income.shape[0]*.8) 
train = income.iloc[:int(train_max_row)]
test = income.iloc[int(train_max_row):]

from sklearn.tree import DecisionTreeClassifier
from sklearn.metrics import roc_auc_score
#得到决策树模型
clf = DecisionTreeClassifier(random_state=10)
#训练它
clf.fit(train[columns],train["high_income"])

prediction=clf.predict(test[columns])
error=roc_auc_score(test["high_income"],prediction)
print error

1.0

prediction=clf.predict(train[columns])
error=roc_auc_score(train["high_income"],prediction)
print error

1.0

#如果训练集和测试集结果相差太大，很有可能是发生了棍和的现象
#解决办法：1.减枝  2.使用多颗树来训练 3.使用限制条件，使深度不至于太深
clf = DecisionTreeClassifier(random_state=10)
#指定最小切分的节点树，也就是说，当前节点如果小于5就不继续切分了
clf = DecisionTreeClassifier(min_samples_split=5,random_state=10)
clf.fit(train[columns],train["high_income"])
prediction=clf.predict(test[columns])
error=roc_auc_score(test["high_income"],prediction)
print error

1.0

clf = DecisionTreeClassifier(random_state=10)
#指定树的最大深度
clf = DecisionTreeClassifier(min_samples_split=13,max_depth=7,random_state=1)
clf.fit(train[columns],train["high_income"])
prediction=clf.predict(test[columns])
error=roc_auc_score(test["high_income"],prediction)
print error

1.0