贝叶斯分类实例（Kaggle比赛之『旧金山犯罪分类预测』）

这里我参照的是http://blog.csdn.net/longxinchen_ml/article/details/50629613
这篇博客。

import pandas as pd
import numpy as np


###############读入数据#################
#用pandas载入csv训练数据，并解析第一列为日期格式
train=pd.read_csv('F://python3.5//Machine Learning//tensorflow//bynet//sf_crime_data//train.csv', parse_dates = ['Dates'])
test=pd.read_csv('F://python3.5//Machine Learning//tensorflow//bynet//sf_crime_data//test.csv', parse_dates = ['Dates'])

#print(train[:5])

#################数据和特征处理##########
import pandas as pd
import numpy as np
from sklearn.cross_validation import train_test_split
from sklearn import preprocessing

#用LabelEncoder对不同的犯罪类型编号
leCrime = preprocessing.LabelEncoder()      
crime = leCrime.fit_transform(train.Category)

#因子化星期几，街区，小时等特征
days = pd.get_dummies(train.DayOfWeek)

district = pd.get_dummies(train.PdDistrict)

hour = train.Dates.dt.hour   #######获取日期中的时间

hour = pd.get_dummies(hour)     
# print(hour[:20])

#组合特征
trainData = pd.concat([hour, days, district], axis=1)
trainData['crime']=crime
# print(trainData[:20])


#对于测试数据做同样的处理
days = pd.get_dummies(test.DayOfWeek)
district = pd.get_dummies(test.PdDistrict)

hour = test.Dates.dt.hour
hour = pd.get_dummies(hour)

testData = pd.concat([hour, days, district], axis=1)


# print(testData[:20])

####################################建模############
'''快速地筛出一部分重要的特征，搭建一个baseline系统，再考虑步步优化。比如我们这里简单一点
,就只取星期几和街区作为分类器输入特征，用scikit-learn中的train_test_split函数拿到训练集和交叉验证集
，用朴素贝叶斯和逻辑回归都建立模型，对比一下它们的表现'''
from sklearn.cross_validation import train_test_split
from sklearn import preprocessing
from sklearn.metrics import log_loss
from sklearn.naive_bayes import BernoulliNB
from sklearn.linear_model import LogisticRegression
import time

""""# 只取星期几和街区作为分类器输入特征
features = ['Monday', 'Tuesday', 'Wednesday', 'Thursday', 'Friday', 'Saturday', 'Sunday', 'BAYVIEW', 'CENTRAL', 'INGLESIDE', 'MISSION',
 'NORTHERN', 'PARK', 'RICHMOND', 'SOUTHERN', 'TARAVAL', 'TENDERLOIN']

# 分割训练集(3/5)和测试集(2/5)
training, validation = train_test_split(trainData, train_size=.60)

# 朴素贝叶斯建模，计算log_loss
model = BernoulliNB()
nbStart = time.time()
model.fit(training[features], training['crime'])
nbCostTime = time.time() - nbStart
predicted = np.array(model.predict_proba(validation[features]))
print ("朴素贝叶斯建模耗时 %f 秒" %(nbCostTime))
print ("朴素贝叶斯log损失为 %f" %(log_loss(validation['crime'], predicted)))


#逻辑回归建模，计算log_loss
model = LogisticRegression(C=.01)
lrStart= time.time()
model.fit(training[features], training['crime'])
lrCostTime = time.time() - lrStart
predicted = np.array(model.predict_proba(validation[features]))
log_loss(validation['crime'], predicted)
print("逻辑回归建模耗时 %f 秒" %(lrCostTime))
print ("逻辑回归log损失为 %f" %(log_loss(validation['crime'], predicted)))"""


'''犯罪类型可能和犯罪事件发生的小时时间点相关，我们加入小时时间点特征再次建模'''

# 添加犯罪的小时时间点作为特征
features = ['Friday', 'Monday', 'Saturday', 'Sunday', 'Thursday', 'Tuesday',
'Wednesday', 'BAYVIEW', 'CENTRAL', 'INGLESIDE', 'MISSION',
'NORTHERN', 'PARK', 'RICHMOND', 'SOUTHERN', 'TARAVAL', 'TENDERLOIN']

hourFea = [x for x in range(0,24)]
features = features + hourFea

print(features)


# 分割训练集(3/5)和测试集(2/5)
training, validation = train_test_split(trainData, train_size=.60)

# 朴素贝叶斯建模，计算log_loss
model = BernoulliNB()
nbStart = time.time()
model.fit(training[features], training['crime'])
nbCostTime = time.time() - nbStart
predicted = np.array(model.predict_proba(validation[features]))
print( "朴素贝叶斯建模耗时 %f 秒" %(nbCostTime))
print ("朴素贝叶斯log损失为 %f" %(log_loss(validation['crime'], predicted)))

#逻辑回归建模，计算log_loss
model = LogisticRegression(C=.01)
lrStart= time.time()
model.fit(training[features], training['crime'])
lrCostTime = time.time() - lrStart
predicted = np.array(model.predict_proba(validation[features]))
log_loss(validation['crime'], predicted)
print ("逻辑回归建模耗时 %f 秒" %(lrCostTime))
print ("逻辑回归log损失为 %f" %(log_loss(validation['crime'], predicted)))

第一次实验结果：

F:\python3.5\lib\site-packages\sklearn\cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)
朴素贝叶斯建模耗时 0.591034 秒
朴素贝叶斯log损失为 2.616172
逻辑回归建模耗时 65.744761 秒
逻辑回归log损失为 2.623363
[Finished in 80.4s]

第二次实验结果：

F:\python3.5\lib\site-packages\sklearn\cross_validation.py:44: DeprecationWarning: This module was deprecated in version 0.18 in favor of the model_selection module into which all the refactored classes and functions are moved. Also note that the interface of the new CV iterators are different from that of this module. This module will be removed in 0.20.
  "This module will be removed in 0.20.", DeprecationWarning)
['Friday', 'Monday', 'Saturday', 'Sunday', 'Thursday', 'Tuesday', 'Wednesday', 'BAYVIEW', 'CENTRAL', 'INGLESIDE', 'MISSION', 'NORTHERN', 'PARK', 'RICHMOND', 'SOUTHERN', 'TARAVAL', 'TENDERLOIN', 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23]
朴素贝叶斯建模耗时 0.801046 秒
朴素贝叶斯log损失为 2.581688
逻辑回归建模耗时 83.199759 秒
逻辑回归log损失为 2.590946
[Finished in 95.1s]