Python Pandas - 结合选择的行和列子集
若要选择行和列子集,请使用 loc。使用索引运算符,即方括号,并在 loc 中设置条件。
假设 Microsoft Excel 中打开的 CSV 文件的内容如下 −

首先,将数据从 CSV 文件加载到 Pandas 数据框中 −
dataFrame = pd.read_csv("C:\Users\amit_\Desktop\SalesData.csv")结合选择行和列子集。右侧列显示要显示的列,此处为汽车列 −
dataFrame.loc[dataFrame["Units"] > 100, "Car"]
示例
代码如下 −
import pandas as pd
# Load data from a CSV file into a Pandas DataFrame:
dataFrame = pd.read_csv("C:\Users\amit_\Desktop\SalesData.csv")
print("\nReading the CSV file...\n",dataFrame)
# selecting a subset of rows
print("\nSelect cars with Units more than 100: \n",dataFrame[dataFrame["Units"] > 100])
# displaying only two columns
res = dataFrame[['Reg_Price','Units']];
print("\nDisplaying only two columns : \n",res)
# Select a subset of rows and columns combined
# Right column displays the column you want to display i.e. Cars column here
res2 = dataFrame.loc[dataFrame["Units"] > 100, "Car"]
# display subset
print("\nSubset...\n",res2)输出
这将生成以下输出 −
Reading the CSV file... Car Reg_Price Units 0 BMW 2500 100 1 Lexus 3500 80 2 Audi 2500 120 3 Jaguar 2000 70 4 Mustang 2500 110 Select cars with Units more than 100: Car Reg_Price Units 2 Audi 2500 120 4 Mustang 2500 110 Displaying only two columns : Reg_Price Units 0 2500 100 1 3500 80 2 2500 120 3 2000 70 4 2500 110 Subset... 2 Audi 4 Mustang Name: Car, dtype: object
广告
数据结构
网络
关系数据库管理系统
操作系统
Java
iOS
HTML
CSS
Android
Python
C 编程
C++
C#
MongoDB
MySQL
Javascript
PHP