def feture_select_RLR():
data_x, data_y, names = get_data()
rlr = RLR()
rlr.fit(data_x, data_y)
return sorted(zip(names, map(lambda x: round(x, 4), rlr.scores_)),
key=lambda x: x[1],
reverse=True)
def get_support_fields(X,Y):
'''
Function for getting support fields
'''
rlr = RLR() #建立随机逻辑回归模型，筛选变量
rlr.fit(X, Y) #训练模型
rlr.get_support() #获取特征筛选结果，也可以通过.scores_方法获取各个特征的分数
print rlr.scores_
print(u'有效特征为：%s' % (','.join(data.columns[rlr.get_support()])).decode('utf-8'))
X = data[data.columns[rlr.get_support()]].as_matrix() #筛选好特征
return X
def programmer_1():
filename = "data/bankloan.xls"
data = pd.read_excel(filename)
x = data.iloc[:, :8].as_matrix()
y = data.iloc[:, 8].as_matrix()
rlr = RLR()
rlr.fit(x, y)
rlr_support = rlr.get_support()
support_col = data.drop('违约', axis=1).columns[rlr_support]
print(
"rlr_support_columns: {columns}".format(columns=','.join(support_col)))
x = data[support_col].as_matrix()
lr = LR()
lr.fit(x, y)
print("lr: {score}".format(score=lr.score(x, y)))
def data_proc(self):
self.load_data()
# iloc，完全基于位置的索引,[]中的第一个值是从第几行到第几行，第二个是从第几列到第几列
x = self.data.iloc[:, :8].as_matrix()
y = self.data.iloc[:, 8].as_matrix()
#先使用随机变量模型进行属性的筛选
rlr = RLR()
rlr.fit(x, y) #训练模型
rlr.get_support() #获取特征筛选结果，也可以通过.scores获得各个特征的分数
print("有效特征为%s" % ','.join(self.data.columns[rlr.get_support()]))
x = self.data[data.columns[rlr.get_support()]].as_matrix() #筛选之后的特征
rlr.get_support()
lr = LR(class_weight={
0: 0.9,
1: 0.1
}) # 分类权重，避免误分类代价比较高时使用，class_weight='balanced'自行处理，或者像代码中那样设置
#lr.fit(x, y,sample_weight=[1,2,3,5,4,9,8,10])
lr.fit(x, y, sample_weight=[1, 2, 3, 5, 4]) #样本权重，设置每一行数据的重要性，一行数据一个值
result = lr.predict([[24, 2, 2, 0, 28, 17.3, 1.79, 3.06]])
print('模型的正确率是：%s,预测结果是 %d' % (lr.score(x, y), result))
def programmer_1():
# 参数初始化
filename = r'bankloan.xls'
data = pd.read_excel(filename)
x = data.iloc[:, :8].as_matrix() # 使用pandas读取文件 就可以不用管label column标签
y = data.iloc[:, 8].as_matrix()
rlr = RLR() # 建立随机逻辑回归模型，进行特征选择和变量筛选
rlr.fit(x, y) # 训练模型
egeList = rlr.get_support() # 获取筛选后的特征
egeList = np.append(
egeList, False) # 往numpy数组中 添加一个False元素 使用np.append(array,ele)方法
print("rlr.get_support():")
print(egeList)
print(u'随机逻辑回归模型特征选择结束！！！')
print(u'有效特征为：%s' % ','.join(data.columns[egeList]))
x = data[data.columns[egeList]].as_matrix() # 筛选好特征值
lr = LR() # 建立逻辑回归模型
lr.fit(x, y) # 用筛选后的特征进行训练
print(u'逻辑回归训练模型结束！！！')
print(u'模型的平均正确率：%s' % lr.score(x, y)) # 给出模型的平均正确率，本例为81.4%
def stable_select(df, y, rd_reg_columns, threshold=0.2, model='rlr'):
X = df.loc[:, rd_reg_columns]
Y = df[y]
if model == 'rlr':
rlr = RLR(scaling=0.5, sample_fraction=0.75, n_resampling=300, selection_threshold=threshold) # 随机逻辑回归
rlr.fit(X, Y)
scores = rlr.scores_
elif model == 'rls':
rls = RLS(scaling=0.5, sample_fraction=0.75, n_resampling=300, selection_threshold=threshold) # 随机Lasso回归
rls.fit(X, Y)
scores = rls.scores_
elif model == 'rfr':
rf = RFR()
rf.fit(X, Y)
scores = rf.feature_importances_
else:
pass
result = pd.Series(dict(zip(X.columns, scores))).rename('score').sort_values(ascending=False)
plt.figure(figsize=(20, 10))
result.plot.barh(title='Feature Importances', color='lightblue')
plt.ylabel('Feature Importance Score')
return result
def logistic(X_train, X_test, y_train, y_test):
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
#特征工程
rlr = RLR()
rlr.fit(X_train, y_train)
print(rlr.get_support())
x = X_train[X_train.columns[rlr.get_support()]].as_matrix()
x_test = X_test[X_test.columns[rlr.get_support()]].as_matrix()
'''
x=X_train
x_test=X_test
'''
#逻辑回归
lr = LR()
lr.fit(x, y_train)
pred_prob_train = lr.predict_proba(x)
pred_prob = lr.predict_proba(x_test)
print('logistic')
predicts = lr.predict(x_test)
metrics_result(y_test, predicts)
return pred_prob, pred_prob_train
def logistic_regression():
# 参数初始化
filename = SRC_PATH + '/data/bankloan.xls'
data = pd.read_excel(filename)
print data.head()
print data.tail()
x = data.iloc[:, :8].as_matrix()
y = data.iloc[:, 8].as_matrix()
print x, y
rlr = RLR() # 建立随机逻辑回归模型，筛选变量
rlr.fit(x, y) # 训练模型
rlr.get_support() # 获取特征筛选结果，也可以通过.scores_方法获取各个特征的分数
print(u'通过随机逻辑回归模型筛选特征结束。')
# print(u'有效特征为：%s' % ','.join(data.columns[rlr.get_support()]))
# x = data[data.columns[rlr.get_support()]].as_matrix() # 筛选好特征
lr = LR() # 建立逻辑货柜模型
lr.fit(x, y) # 用筛选后的特征数据来训练模型
print(u'逻辑回归模型训练结束。')
print(u'模型的平均正确率为：%s' % lr.score(x, y)) # 给出模型的平均正确率，本例为81.4%
import pandas as pd
from sklearn.linear_model import RandomizedLogisticRegression as RLR
from sklearn.linear_model import LogisticRegression as LR
#读取数据
data = pd.read_csv("C:/Users/T/Desktop/python视频/luqu.csv")
x = data.iloc[:, 1:4].as_matrix()
y = data.iloc[:, :1].as_matrix()
#随机Logistic模型，用于筛选变量
f1 = RLR()
f1.fit(x, y)
f1.get_support() #筛选出的变量
#Logistic模型
f2 = LR()
f2.fit(x, y)
f2.score(x, y) #准确率
# coding: utf-8
# ���ع� �Զ���ģ
import pandas as pd
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
filename = '../data/bankloan.xls'
data = pd.read_excel(filename)
x = data.iloc[:, :8].as_matrix()
y = data.iloc[:, 8].as_matrix()
rlr = RLR(selection_threshold=0.5) # 建立模型
rlr.fit(x, y) # 训练
rlr.get_support() # 获取特征筛选结果
print('通过随机逻辑回归模型筛选结束')
print('有效特征：%s' % ','.join(data.columns[rlr.get_support(indices=True)]))
x = data[data.columns[rlr.get_support(indices=True)]].as_matrix() # 筛选好特征
lr = LR()
lr.fit(x, y)
print(u'逻辑回归模型训练结束')
print(u'模型平均正确率：%s' % lr.score(x, y)) # 本例 ：81.4%
import pandas as pd
import numpy as np
import matplotlib.pylab as plt
from scipy import ndimage
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
data = pd.read_csv("luqu.csv")
x = data.iloc[:, 1:5]
y = data.iloc[:, 0:1]
#####首先进行特征筛选#####
rlr = RLR() #建立随机逻辑回归模型，进行特征选择和变量筛选
rlr.fit(x, y) #训练模型
egeList = rlr.get_support() #获取筛选后的特征
print("rlr.get_support():")
print(egeList)
print('有效特征为：' + str(data.iloc[:, 1:5].columns[egeList].values))
print('随机逻辑回归模型特征选择结束！！！')
x = data[data.iloc[:, 1:5].columns[egeList]] #筛选好特征值，排除无效的特征值
#####确定逻辑回归参数#####
# 为确定参数C值，采用隔点搜索
# 用logspace先产生一组非常微小的数
from sklearn.model_selection import cross_val_score
lr = LR()
Pars = np.logspace(-4, 4, num=20)
# -*- coding:utf-8 -*- import pandas as pd filename = '../data/bankloan.xls' data = pd.read_excel(filename) x = data.iloc[:,:8].as_matrix() y = data.iloc[:,8].as_matrix() from sklearn.linear_model import LogisticRegression as LR from sklearn.linear_model import RandomizedLogisticRegression as RLR rlr = RLR() #建立随机逻辑回归模型,复筛选变量 rlr.fit(x, y) #训练模型 rlr.get_support() #获取特征筛选变量 print(u'有效特征为:%s' % ','.join(data.columns[rlr.get_support()])) x = data[data.columns[rlr.get_support()]].as_matrix() #筛选锟斤拷锟斤拷锟斤拷 lr = LR() #建立逻辑回归模型 lr.fit(x, y) #训练模型 print(u'模型的平均正确率:%s' % lr.score(x, y))
#-*- coding: utf-8 -*-
#逻辑回归 自动建模
import pandas as pd
#参数初始化
filename = 'pdata.xls'
data = pd.read_excel(filename)
x = data.iloc[:, :3].as_matrix()
y = data.iloc[:, 3].as_matrix()
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR #Lasso/f_rgression
rlr = RLR(selection_threshold=0.25) #建立随机逻辑回归模型，筛选变量，默认阈值0.25，
rlr.fit(x, y) #训练模型
rlr.get_support() #获取特征筛选结果，也可以通过.scores_方法获取各个特征的分数
print('通过随机逻辑回归模型筛选特征结束。')
print rlr.scores_
print('%s' % ','.join(data.columns[rlr.get_support()]))
x = data[data.columns[rlr.get_support()]].as_matrix() #筛选好特征
lr = LR() #建立逻辑货柜模型
lr.fit(x, y) #用筛选后的特征数据来训练模型
print('逻辑回归模型训练结束。')
print('correct_point：%s' % lr.score(x, y)) #给出模型的平均正确率
# -*- coding: utf-8 -*-
# 时间 : 2018/9/15 13:38
# 作者 : xcl
import pandas as pd
filename = 'C:/Users/Administrator/Desktop/bankloan.xls'
data = pd.read_excel(filename)#返回值是DataFrame类型
x = data.iloc[:,:8].values#行全选，列选下标0-7
y = data.iloc[:,8].values#行全选，列选下标8
#使用稳定性选择方法中的随机逻辑回归进行特征筛选，利用筛选后的特征建立逻辑回归模型，输出平均正确率
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
rlr = RLR(selection_threshold=0.25i) #建立随机逻辑回归模型，筛选变量
rlr.fit(x, y) #训练模型
#print(rlr.get_support()) #获取特征筛选结果，也可以通过.scores_方法获取各个特征的分数
#print(rlr.scores_)
#print('通过随机逻辑回归模型筛选特征结束')
print('有效特征:' , ','.join(data.columns[rlr.get_support(indices=True)]))
#print(data.columns[rlr.get_support(indices=True)])
#data.columns[rlr.get_support()]返回的是筛选后的列名,是一个迭代器
#S.join(iterable) 将iterable里面的元素用S连起来，S就是分隔符
x = data[data.columns[rlr.get_support(indices=True)]].values
lr = LR() #建立逻辑回归模型
lr.fit(x, y) #用筛选后的特征数据来训练模型
#print('逻辑回归模型训练结束')
#-*- coding: utf-8 -*- #逻辑回归 自动建模 import pandas as pd from sklearn.linear_model import LogisticRegression as LR from sklearn.linear_model import RandomizedLogisticRegression as RLR #参数初始化 filename = '../data/bankloan.xls' data = pd.read_excel(filename) x = data.iloc[:,:8].as_matrix()#8个属性 y = data.iloc[:,8].as_matrix()#第九列 结果标签 #稳定性选择方法 挑选特征 rlr = RLR(selection_threshold=0.5) #建立随机逻辑回归模型，筛选变量 特征筛选用了默认阈值0.25 rlr.fit(x, y) #训练模型 rlr.get_support() #获取特征筛选结果 print(u'通过随机逻辑回归模型筛选特征结束。') print(u'有效特征为：%s' % ','.join(data.columns[rlr.get_support()])) x = data[data.columns[rlr.get_support()]].as_matrix() #筛选好特征,重新训练模型 lr = LR() #建立逻辑货柜模型 lr.fit(x, y) #用筛选后的特征数据来训练模型 print(u'逻辑回归模型训练结束。') print(u'模型的平均正确率为：%s' % lr.score(x, y))
#-*- coding:utf-8 -*-
import pandas as pd
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
filename = '../../data/5-/bankloan.xls'
data = pd.read_excel(filename)
x = data.iloc[:, :8].as_matrix()
y = data.iloc[:, 8].as_matrix()
#开始进行属性规约。
rlr = RLR() #建立随机逻辑货柜模型，筛选变量
rlr.fit(x, y)
m = list(rlr.get_support()) #获取特征筛选结果
print(u'适合的特征选出来了。如下：')
n = []
for i in range(len(m)):
if m[i] == True:
n.append(data.columns[i])
print(data.columns[i])
#columns内参数应为int型.
x = data[n].as_matrix() #将筛选好特征的数据列出
#以上进行了数据预处理中的属性规约。
#建立逻辑回归模型进行测试。
lr = LR()
lr.fit(x, y)
print(u'模型训练结束')
print(u'模型的平均正确率为：%s' % lr.score(x, y))
# 评判结果报告
from sklearn.metrics import classification_report
def my_classification_report(y_true, y_pred):
from sklearn.metrics import classification_report
print( "classification_report(left: labels):")
print (classification_report(y_true, y_pred))
my_classification_report(y_test,y_score_pre)
#筛选特征RLR
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
Rlogis_reg = RLR(random_state=10) #筛选变量
Rlogis_reg.fit(X_train, y_train)
selected_col =list(Rlogis_reg.get_support())
selected_col.append(False)
print(u"通过随机逻辑回归模型筛选特征结束")
print(u"有效特征为：%s" % ",".join(smote_resampled.columns[selected_col]))
X1_train = pd.DataFrame(X_train,columns=names)
X1_test = pd.DataFrame(X_test,columns=names)
selected_feature =smote_resampled.columns[selected_col]
X1_train = X1_train[selected_feature]
X1_test = X1_test[selected_feature] # 筛选好特征
RLR=model(LR,X1_train,X1_test,y_train,y_test,model_name='RLR 个4 特征：')
#决策树
#参数初始化 filename = './data/bankloan.xls' data = pd.read_excel(filename) x = data.iloc[:, :8].as_matrix() y = data.iloc[:, 8].as_matrix() from sklearn.linear_model import LogisticRegression as LR from sklearn.linear_model import RandomizedLogisticRegression as RLR lr = LR() #建立逻辑货柜模型 lr.fit(x, y) #用筛选后的特征数据来训练模型 print(u'逻辑回归模型训练结束。') print(u'未经过筛选特性模型的平均正确率为：%s' % lr.score(x, y)) #建立随机逻辑回归模型 rlr = RLR() #帅选变量 rlr.fit(x, y) #rlr.get_support() #获取特征筛选结果，也可以通过.scores_方法获取各个特征的分数 selected_col = numpy.append(rlr.get_support(),[False]) print(u"通过随机逻辑回归模型筛选特征结束") print(u"有效特征为：%s" % ",".join(data.columns[selected_col])) x = data[data.columns[selected_col]].as_matrix() # 筛选好特征 lr = LR() #建立逻辑货柜模型 lr.fit(x, y) #用筛选后的特征数据来训练模型 print(u'逻辑回归模型训练结束。') print(u'模型的平均正确率为：%s' % lr.score(x, y)) #给出模型的平均正确率，本例为81.4%
import pandas as pd
from pandas import DataFrame as df
from sklearn.ensemble import RandomForestRegressor
from sklearn.ensemble import RandomForestClassifier
dataFile = r'F:\pycharm_workspace\myML_DM_Test\resource\python_practice_Data_Analy_Min\chapter5\chapter5\demo\data\bankloan.xls'
data = pd.read_excel(dataFile)
df_data = df(data)
print(data)
print("DF: \n", df_data)
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
x = data.iloc[:, :8].as_matrix()
y = data.iloc[:, 8].as_matrix()
print("X \n", x)
print("Y \n", y)
rlr = RLR() #建立随机逻辑回归模型，筛选变量
rlr.fit(x, y) #训练模型
result.sort()
print(result[-1])
model = LR(
penalty='l2',
tol=0.0001,
C=.005,
solver='liblinear',
max_iter=500,
)
model.fit(df, label)
print(features, model.coef_)
model2 = RLR(C=1,
scaling=0.5,
sample_fraction=0.6,
selection_threshold=0.3,
n_resampling=100)
model2.fit(df, label)
print(model2.get_support())
# dt ------------------------------------------
d = [1, 2, 3, 4, 5, 10, 15, 20, 25]
m = ['gini', 'entropy']
result = [
bin_cv(DTC(criterion='entropy', max_depth=depth), df, label).mean()
for depth in d
]
print(result)
pd.Series(result).plot()
#-*- coding:utf-8 -*-
# Peishichao
import pandas as pd
filename = '../data/bankloan.xls'
data = pd.read_excel(filename)
x = data.iloc[:, :8].as_matrix()
y = data.iloc[:, 8].as_matrix()
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
rlr = RLR()
rlr.fit(x, y)
rlr.get_support()
print(rlr.get_support())
print('end')
#print('Feature: %s ' % ','.join(data.columns[rlr.get_support()]))
x = data[data.columns[rlr.get_support()]].as_matrix()
print(x)
lr = LR()
lr.fit(x, y)
print('end')
print('accur: %s' % lr.score(x, y))
#-*- coding: utf-8 -*-
import pandas as pd
import os
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
# init parameters
# os.getcwd()
filename = '../Datasets/bankloan.csv'
data = pd.read_csv(filename)
x = data.iloc[:, :8].as_matrix() # regressors
y = data.iloc[:, 8].as_matrix() # labels
features_columns = data.columns[:len(data.columns) - 1]
rlr = RLR() # create a randomized LogisticRegression, feature selection
rlr.fit(x, y) # training model
print(features_columns)
print(rlr.get_support()) # get feature selection result.
print(rlr.scores_) # get each feature score
print(u'RandomizedLogisticRegression feature selection finished.')
print(u'The effective features are:\n\t %s' %
', '.join(features_columns[rlr.get_support()]))
x = data[features_columns[rlr.get_support()]].as_matrix() # selected features
lr = LR() # create logistic regression model
lr.fit(x, y) # using effective features training model
print(u'Accuracy：%s' % lr.score(x, y))
import pandas as pd
from sklearn.linear_model import LogisticRegression as LR
from sklearn.linear_model import RandomizedLogisticRegression as RLR
lessonPath = 'E:\\BaiduNetdiskDownload\\sourceCode\\week8\\lesson2.csv'
luquPath = 'E:\\BaiduNetdiskDownload\\sourceCode\\week8\\luqu2.csv'
dataLuqu = pd.read_csv(luquPath)
# 特征
x = dataLuqu.iloc[:, 1:4].as_matrix()
y = dataLuqu.iloc[:, 0:1].as_matrix()
r1 = RLR()
r1.fit(x, y)
# 特征筛选
r1.get_support()
t = dataLuqu.columns[r1.get_support()].as_matrix()
r2 = LR()
r2.fit(t, y)
print('训练结束')
print('模型正确率: ' + str(r2.score(x, y)))
def main():
import read_data
import data_standard
import numpy as np
import pandas as pd
'''
##############################################################数据读入与整理##################################################################################
path2014='E:\Master\PPDAMcode\AIR_project\hangzhou.xls'
path2015='E:\Master\PPDAMcode\AIR_project\hangzhou2015.xls'
path2016='E:\Master\PPDAMcode\AIR_project\hangzhou2016.xls'
hangzhou_data=[]
##################################################################################
data=read_data.read_data_xls(path2014)
for i in range(len(data)-1):
if data.iloc[i+1,7]=='优':
data.iloc[i+1,7]=1.0
elif data.iloc[i+1,7]=='良好':
data.iloc[i+1,7]=2.0
elif data.iloc[i+1,7]=='轻度污染':
data.iloc[i+1,7]=3.0
elif data.iloc[i+1,7]=='中度污染':
data.iloc[i+1,7]=4.0
elif data.iloc[i+1,7]=='重度污染':
data.iloc[i+1,7]=5.0
#for i in range(6):
#data.iloc[1:,1+i]=data_standard.data_standard(data.iloc[1:,1+i])
hangzhou_data.append(data)
#################################################################################
data=read_data.read_data_xls(path2015)
for i in range(len(data)-1):
if data.iloc[i+1,7]=='优':
data.iloc[i+1,7]=1.0
elif data.iloc[i+1,7]=='良好':
data.iloc[i+1,7]=2.0
elif data.iloc[i+1,7]=='轻度污染':
data.iloc[i+1,7]=3.0
elif data.iloc[i+1,7]=='中度污染':
data.iloc[i+1,7]=4.0
elif data.iloc[i+1,7]=='重度污染':
data.iloc[i+1,7]=5.0
#for i in range(6):
#data.iloc[1:,1+i]=data_standard.data_standard(data.iloc[1:,1+i])
hangzhou_data.append(data)
##################################################################################
data=read_data.read_data_xls(path2016)
for i in range(len(data)-1):
if data.iloc[i+1,7]=='优':
data.iloc[i+1,7]=1.0
elif data.iloc[i+1,7]=='良好':
data.iloc[i+1,7]=2.0
elif data.iloc[i+1,7]=='轻度污染':
data.iloc[i+1,7]=3.0
elif data.iloc[i+1,7]=='中度污染':
data.iloc[i+1,7]=4.0
elif data.iloc[i+1,7]=='重度污染':
data.iloc[i+1,7]=5.0
#for i in range(6):
#data.iloc[1:,1+i]=data_standard.data_standard(data.iloc[1:,1+i])
hangzhou_data.append(data)
############################################################################
#print(hangzhou_data)#######2014,2015,2016的所有数据在列表hangzhou_data中存放
all_data=pd.DataFrame(np.linspace(1,1091*8,1091*8).reshape(1091,8))
for i in range(len(hangzhou_data[0])-1):
for j in range(8):
all_data.iloc[i,j]=hangzhou_data[0].iloc[i+1,j+1]
############################################
for i in range(len(hangzhou_data[1])-1):
for j in range(8):
all_data.iloc[i+365,j]=hangzhou_data[1].iloc[i+1,j+1]
###################################################
for i in range(len(hangzhou_data[2])-1):
for j in range(8):
all_data.iloc[i+730,j]=hangzhou_data[2].iloc[i+1,j+1]
#print(all_data)
all_data.to_excel('E:\Master\PPDAMcode\AIR_project\hangzhou_air_alldata.xls')
################################################################################################################################################
'''
############################主成分分析PCA######################################################
import PCA
PCA.PCA(all_data.iloc[:,0:6])
#######################RandomizedLogisticRegression筛选特征##########
from sklearn.linear_model import RandomizedLogisticRegression as RLR
rlr=RLR()
rlr.fit(all_data.iloc[:,0:6].as_matrix(),all_data.iloc[:,6].as_matrix())
print()
print('通过RandomizedLogisticRegression筛选特征：')
print(rlr.get_support())
print()
###############################################################################################
#############################准备好训练数据与测试数据################################################################
all_data=read_data.read_data_xls('E:\Master\PPDAMcode\AIR_project\yuanshi_data.xls')
import random
index=list(range(1091))
test_index = random.sample(index, 90)##test_index is the index of test data随机选出200个样本作为测试样本
train_index=[]
for i in range(1091):
if i not in test_index:
train_index.append(i) ##train_index is the index of train data
data_train=all_data.iloc[train_index,0:6].as_matrix()
classfy_train=all_data.iloc[train_index,6].as_matrix()
AQI_train=all_data.iloc[train_index,7].as_matrix()
data_test=all_data.iloc[test_index,0:6].as_matrix()
classfy_test=all_data.iloc[test_index,6].as_matrix()
AQI_test=all_data.iloc[test_index,7].as_matrix()
########################################################################################################################
'''
############################运用DecisionTreeClassifier方法进行训练测试###############################
import DecisionTreeClassifier
DecisionTreeClassifier.DecisionTreeClassifier(data_train,AQI_train,data_test,AQI_test)
######################################################################################################
'''
############################运用RandomForestClassifier方法进行训练测试###############################
import RandomForestClassifier
RandomForestClassifier.RandomForestClassifier(data_train,AQI_train,data_test,AQI_test)
######################################################################################################
'''
############################运用SVM方法进行训练测试###############################
import SVM
SVM.SVM(data_train,classfy_train,data_test,classfy_test)
######################################################################################################
'''
'''
############################运用keras方法进行训练测试###############################
import Sequential
Sequential.Sequential(data_train,classfy_train,data_test,classfy_test)
######################################################################################################
'''
'''