我的代码中 pandas 到 csv 有什么问题？

Question

我是运行我正在做的一个项目的代码，目的是寻找迪斯尼乐园等待时间的模式：

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt
df_pirates_all = pd.read_csv(
    "https://cdn.touringplans.com/datasets/pirates_of_caribbean_dlr.csv",usecols=['date','datetime','SPOSTMIN'],
    parse_dates=['date', 'datetime'], 
)
df_pirates_all['ride'] = 'pirates'
df_pirates_all['open'] = ~((df_pirates_all['SPOSTMIN'] == -999))

df_pirates = df_pirates_all.set_index('datetime').sort_index()
df_pirates = df_pirates.loc['2017-01-01 06:00':'2017-02-01 00:00']
df_pirates = df_pirates.resample('15Min').ffill()


df_star_tours_all = pd.read_csv(
    "https://cdn.touringplans.com/datasets/star_tours_dlr.csv", usecols=['date','datetime','SPOSTMIN'],
    parse_dates=['date', 'datetime']
)
df_star_tours_all['ride'] = 'star_tours'
df_star_tours_all['open'] = ~((df_star_tours_all['SPOSTMIN'] == -999))

df_star_tours = df_star_tours_all.set_index('datetime').sort_index()
df_star_tours = df_star_tours.loc['2017-01-01 06:00':'2017-02-01 00:00']
df_star_tours = df_star_tours.resample('15Min').ffill()

df_space_all = pd.read_csv(
    "https://cdn.touringplans.com/datasets/space_mountain_dlr.csv", usecols=['date','datetime','SPOSTMIN'], 
    parse_dates=['date', 'datetime']
)
df_space_all['ride'] = 'space'
df_space_all['open'] = ~((df_space_all['SPOSTMIN'] == -999))

df_space = df_space_all.set_index('datetime').sort_index()
df_space = df_space.loc['2017-01-01 06:00':'2017-02-01 00:00']
df_space = df_space.resample('15Min').ffill()


all_data = pd.concat([df_pirates, df_star_tours, df_space]).reset_index()
all_data = (
    all_data
        # Drop any "NaN" values in the column 'ride'
        .dropna(subset=['ride', ])
        # Make datetime and ride a "Multi-Index"
        .set_index(['datetime', 'ride'])
        # Choose the column 'SPOSTMIN'
        ['SPOSTMIN']
        # Take the last index ('ride') and rotate to become column names
        .unstack()
)
# print (all_data)

for month, group in all_data.groupby(pd.Grouper(freq='M')):
    with pd.ExcelWriter(f'{month}.xlsx') as writer:
        for day, dfsub in group.groupby(pd.Grouper(freq='D')):
            dfsub.to_excel(writer, sheet_name='day')

但是我运行遇到了这个错误

FileCreateError: [Errno 22] Invalid argument: '2017-01-31 00:00:00.xlsx'

并且连接到dfsub.to_excel线。

它大部分已通过评论修复，但是，只有一个 sheet 出现并且它只有最后一天的数据 (1-31-17) 而不是单独的 sheet 1-1-17,1-2-17,等等

Answer 1

对于基于代码的第一个错误，您不关心具体的日期和时间，所以这样做：

with pd.ExcelWriter(f'{month.date()}.xlsx'):

这会将日期时间对象转换为日期对象

你的第二个错误是说你正在尝试创建一个不完全唯一的列作为索引，pandas 不允许。

也许有些字段可以合并或使用另一个字段？

Answer 2

解决问题的方法是更改

中的代码

for month, group in all_data.groupby(pd.Grouper(freq='M')):
    with pd.ExcelWriter(f'{month}.xlsx') as writer:
        for day, dfsub in group.groupby(pd.Grouper(freq='D')):
            dfsub.to_excel(writer, sheet_name='day')

至

for month, group in all_data.groupby(pd.Grouper(freq='M')):
    with pd.ExcelWriter(f'{month.strftime("%B %Y")}.xlsx') as writer:
        for day, dfsub in group.groupby(pd.Grouper(freq='D')):
            dfsub.to_excel(writer,sheet_name=str(day.date()))

根据提出的建议。

我的代码中 pandas 到 csv 有什么问题？

What is the problem with the pandas to csv in my code?

python

numpy

export-to-excel

pandas