📜  在 Pandas 聚合中计数不同

📅  最后修改于: 2022-05-13 01:54:59.179000             🧑  作者: Mango

在 Pandas 聚合中计数不同

在本文中,让我们看看如何在 Pandas 聚合中计算不同。所以为了计算 pandas 聚合中的不同,我们将使用 groupby() 和 add() 方法。

  • 通过...分组(): 此方法用于根据某些标准将数据分组。 Pandas 对象可以在它们的任何轴上拆分。我们可以创建一组类别并将函数应用于这些类别。分组的抽象定义是提供标签到组名的映射
  • agg():此方法用于传递一个函数或函数列表,以分别应用于系列甚至系列的每个元素。在函数列表的情况下,agg() 方法返回多个结果。

以下是一些描述如何在 Pandas 聚合中计算不同值的示例:

示例 1:

Python
# import module
import pandas as pd
import numpy as np
  
# create Data frame
df = pd.DataFrame({'Video_Upload_Date': ['2020-01-17',
                                         '2020-01-17',
                                         '2020-01-19',
                                         '2020-01-19',
                                         '2020-01-19'],
                   'Viewer_Id': ['031', '031', '032',
                                 '032', '032'],
                   'Watch_Time': [34, 43, 43, 41, 40]})
  
# print original Dataframe
print(df)
  
# let's Count distinct in Pandas aggregation
df = df.groupby("Video_Upload_Date").agg(
    {"Watch_Time": np.sum, "Viewer_Id": pd.Series.nunique})
  
# print final output
print(df)


Python
# import module
import pandas as pd
import numpy as np
  
# create Data frame
df = pd.DataFrame({'Order Date': ['2021-02-22',
                                  '2021-02-22',
                                  '2021-02-22',
                                  '2021-02-24',
                                  '2021-02-24'],
                   'Product Id': ['021', '021',
                                  '022', '022', '022'],
                   'Order Quantity': [23, 22, 22,
                                      45, 10]})
  
# print original Dataframe
print(df)
  
# let's Count distinct in Pandas aggregation
df = df.groupby("Order Date").agg({"Order Quantity": np.sum,
                                   "Product Id": pd.Series.nunique})
  
# print final output
print(df)


输出:



示例 2:

Python

# import module
import pandas as pd
import numpy as np
  
# create Data frame
df = pd.DataFrame({'Order Date': ['2021-02-22',
                                  '2021-02-22',
                                  '2021-02-22',
                                  '2021-02-24',
                                  '2021-02-24'],
                   'Product Id': ['021', '021',
                                  '022', '022', '022'],
                   'Order Quantity': [23, 22, 22,
                                      45, 10]})
  
# print original Dataframe
print(df)
  
# let's Count distinct in Pandas aggregation
df = df.groupby("Order Date").agg({"Order Quantity": np.sum,
                                   "Product Id": pd.Series.nunique})
  
# print final output
print(df)

输出: