Question

我有两列数据（sample data）我想计算每个工作日的总用户数。

例如，我希望我的输出像这样（dict / list会做什么）：

周一：25日，星期二：30，周三：45，星期四：50，周五：24，星期六：22，周日：21

这是我的尝试：

def rider_ship (filename):
    with open('./data/Washington-2016-Summary.csv','r') as f_in:

        Sdict = []
        Cdict = []
        reader = csv.DictReader(f_in)
        for row in reader:
            if row['user_type']=="Subscriber": 
                if row['day_of_week'] in Sdict:
                    Sdict[row['day_of_week']]+=1
                else: 
                    Sdict [row['day_of_week']] = row['day_of_week']
            else:
                if row ['day_of_week'] in Cdict:
                    Cdict[row['day_of_week']] +=1
                else: 
                     Cdict[row['day_of_week']] = row['day_of_week']

        return Sdict, Cdict

        print (Sdict)
        print (Cdict)

t= rider_ship ('./data/Washington-2016-Summary.csv')
print (t)

TypeError :: list indices必须是整数或切片，而不是str

Answer 1

使用熊猫怎么样？

让我们首先使用io库创建一个类文件对象：

import io
s = u"""day_of_week,user_type
Monday,subscriber
Tuesday,customer
Tuesday,subscriber
Tuesday,subscriber"""
file = io.StringIO(s)

现在到实际代码：

import pandas as pd
df = pd.read_csv(file) # "path/to/file.csv"

Sdict = df[df["user_type"] == "subscriber"]["day_of_week"].value_counts().to_dict()
Cdict = df[df["user_type"] == "customer"]["day_of_week"].value_counts().to_dict()

现在我们有：

Sdict = {'Tuesday'：2，'Monday'：1}

Cdict = {'Tuesday'：1}

我的代码没有产生任何输出 - Python

1 个答案: