if command=='3':
data_name=input('请输入数据的文件名(包含路径)，缺省为 .\\data\\data.xlsx. 请输入:')
if not data_name:
data_name='.\\data\\data.xlsx'
try:
data=rpt.read_data(data_name)
print('已成功导入data.')
except Exception as e:
print(e)
print('data导入失败, 请检查')
continue
code_name=input('请输入code的文件名(包含路径)，缺省为 .\\data\\code.xlsx. 请输入:')
if not code_name:
code_name='.\\data\\code.xlsx'
try:
code=rpt.read_code(code_name)
print('已成功导入code.')
except Exception as e:
print(e)
print('code导入失败, 请检查')
continue
cross_qlist=list(sorted(code,key=lambda c: int(re.findall('\d+',c)[0])))
print('-'*20+'题目数:{}个,样本数：{}个'.format(len(code),len(data))+'-'*20)
print('题目编码情况如下......\n')
for k in cross_qlist:
print('{key}: {c}'.format(key=k,c=code[k]['content']))
time.sleep(0.1)
break
if command=='4':
try:
import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
import report as rpt
reload(rpt)
# 数据d导入
code=rpt.read_code('.\\data\\code.xlsx')
data0=pd.read_excel('.\\data\\data.xlsx',encoding='gbk')
# 数据清晰
data=data0[(data0['Q5']==1)|(data0['Q5'].isnull())]#清楚自己购买但使用不是自己的人
data=data[data[u'来源详情']==u'直接访问']
'''
Q12=data[code['Q12']['qlist']]
Q12.applymap(lambda x:int(x==1))
Q12=Q12.sum()
Q12.rename(index=code['Q12']['code'],inplace=True)
Q12.sort_values(inplace=True)
'''
filename=u'小鲜4真实使用用户1_334'
rpt.summary_chart(data,code,filename=filename)
"""
# 聚类分析
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from scipy import stats
#import seaborn as sns; sns.set()
import report as rpt
from imp import reload
reload(rpt)
# 数据d导入
code = rpt.read_code('code.xlsx')
data = pd.read_excel('data.xlsx')
data['Q21a'] = data[code['Q21']['qlist']].T.max().T
code['Q21a'] = {
'content': '拍照频率',
'qtype': '单选题',
'qlist': ['Q21a'],
'code': {
1: '几乎不使用',
2: '偶尔用',
3: '经常用'
}
}
from sklearn.cluster import KMeans