pandas中DataFrame直接索引就是df[xx]的形式，可以根据标签索引(index、columns参数)或者位置索引选取数据，也可以进行切片操作。DataFrame是2维数据结构，无论行还是列都存在位置索引，默认都是从0开始。

　　我们按照这个df为例来进行学习。

# -*- coding: utf-8 -*-

import pandas as pd

df = pd.DataFrame([['bj','100','200'],['sh','80','60'],['nj','50','30']],columns=['city','sale','rent'],index=['a','b','c'])
print(df)

  city sale rent
a   bj  100  200
b   sh   80   60
c   nj   50   30

　　1、列标签索引选择单列，返回Series

df = pd.DataFrame([['bj','100','200'],['sh','80','60'],['nj','50','30']],columns=['city','sale','rent'],index=['a','b','c'])
print(df['city'])

a    bj
b    sh
c    nj
Name: city, dtype: object

　　2、列标签索引选择单列，返回DataFrame

df = pd.DataFrame([['bj','100','200'],['sh','80','60'],['nj','50','30']],columns=['city','sale','rent'],index=['a','b','c'])
print(df[['city']])

  city
a   bj
b   sh
c   nj

　　3、列标签索引选择多列，返回DataFrame

df = pd.DataFrame([['bj','100','200'],['sh','80','60'],['nj','50','30']],columns=['city','sale','rent'],index=['a','b','c'])
print(df[['city','sale']])

  city sale
a   bj  100
b   sh   80
c   nj   50

　　4、行标签切片(末端包含)选择多行，返回DataFrame

df = pd.DataFrame([['bj','100','200'],['sh','80','60'],['nj','50','30']],columns=['city','sale','rent'],index=['a','b','c'])
print(df['b':'c'])

  city sale rent
b   sh   80   60
c   nj   50   30

　　5、行位置索引切片(末端不包含)选择行，DataFrame

df = pd.DataFrame([['bj','100','200'],['sh','80','60'],['nj','50','30']],columns=['city','sale','rent'],index=['a','b','c'])
print(df[1:2])

  city sale rent
b   sh   80   60

　　6、行布尔索引选择多行

df = pd.DataFrame([['bj','100','200'],['sh','80','60'],['nj','50','30']],columns=['city','sale','rent'],index=['a','b','c'])
print(df[[True,True,False]])

  city sale rent
a   bj  100  200
b   sh   80   60

很赞哦！

python编程网提示：转载请注明来源www.python66.com。
有宝贵意见可添加站长微信(底部)，获取技术资料请到公众号(底部)。同行交流请加群