来自时间序列数据帧的matplotlib [英] matplotlib from time series data frame

问题描述

说我有一个像这样的数据框:

Say I have a data frame like this:

from pandas import DataFrame
example = {'year_month':  [201801,201802,201803,201801,201802,201803],
        'store_id': [101,101,101,102,102,102],
       'tot_employees': [100,200,150,6,7,10],
       'hrs_per_employee': [30,35,20,20,18,15]
        }
df = DataFrame(example,columns=["year_month", "store_id", "tot_employees", "hrs_per_employee"])
df

并且我想为每个store_id堆叠具有不同子图的子图，

and i want to have stacked subplots with a different subplot for each store_id with:

• x轴:year_month
• 第1行:第一名员工
• 情节第2行:每小时 员工

• x axis: year_month
• plots line 1: tot employees
• plot line 2: hrs per employee

使用df.plot()是否可能?我一直无法找到正确的x，y输入来获得我正在寻找的结果.如果没有，还有没有其他选择?预先感谢

is this possible with df.plot()? I haven't been able to find the correct x,y inputs to get the result i'm looking for. if not is there a close alternative? thanks in advance

推荐答案

import pandas as pd
df = pd.DataFrame(example)
df.year_month = pd.to_datetime(df.year_month, format='%Y%m', exact=True)
df.set_index('year_month', drop=True, inplace=True)

for x in df.store_id.unique():
    df[['tot_employees', 'hrs_per_employee']][df.store_id == x].plot(title=f'Store ID: {x}')

df.groupby('store_id').plot(y=['tot_employees', 'hrs_per_employee'])

这篇关于来自时间序列数据帧的matplotlib的文章就介绍到这了，希望我们推荐的答案对大家有所帮助，也希望大家多多支持IT屋！