热搜：编程 Python Microsoft 编程语言 C++

Python3.5 Pandas模块之DataFrame用法实例分析

发布时间：2023-05-25 05:46:56 来源：好代码

.

柴门鸟雀噪，归客千里至。光明顶下来，一路听着阵阵松涛，我们走过西海，飞来石，天海，爬过百步云梯，伴着凤凰松迎客松的多姿到了玉屏峰。玉屏峰犹如一道屏障，山面光滑如镜子，如屏风，还不时有几个斗大的字，说是哪位名人留下的。玉屏峰最险的地方是在一线天，上下达七八十度的落差，台阶狭，仅容一人，台阶窄，仅托一足，百米长，走多了，腿便打抖了，害怕了，还不时地叫着好刺激呀。

本文实例讲述了Python3.5 Pandas模块之DataFrame用法。分享给大家供大家参考，具体如下：

1、DataFrame的创建

（1）通过二维数组方式创建

#!/usr/bin/env python
# -*- coding:utf-8 -*-
# Author:ZhengzhengLiu

import numpy as np
import pandas as pd
from pandas import Series,DataFrame

#1.DataFrame通过二维数组创建
print("======DataFrame直接通过二维数组创建======")
d1 = DataFrame([["a","b","c","d"],[1,2,3,4]])
print(d1)

print("======DataFrame借助array二维数组创建======")
arr = np.array([
  ["jack",78],
  ["lili",86],
  ["amy",97],
  ["tom",100]
])

d2 = DataFrame(arr,index=["01","02","03","04"],columns=["姓名","成绩"])
print(d2)
print("========打印行索引========")
print(d2.index)
print("========打印列索引========")
print(d2.columns)
print("========打印值========")
print(d2.values)

运行结果：

======DataFrame直接通过二维数组创建======
0 1 2 3
0 a b c d
1 1 2 3 4
======DataFrame借助array二维数组创建======
姓名成绩
01 jack 78
02 lili 86
03 amy 97
04 tom 100
========打印行索引========
Index(['01', '02', '03', '04'], dtype='object')
========打印列索引========
Index(['姓名', '成绩'], dtype='object')
========打印值========
[['jack' '78']
['lili' '86']
['amy' '97']
['tom' '100']]

（2）通过字典方式创建

#2.DataFrame通过字典创建，键作为列索引，键值作为数据值，行索引值自动生成

data = {
  "apart":['1101',"1102","1103","1104"],
  "profit":[2000,4000,5000,3500],
  "month":8
}

d3 = DataFrame(data)
print(d3)
print("========行索引========")
print(d3.index)
print("========列索引========")
print(d3.columns)
print("========数据值========")
print(d3.values)

运行结果：

apart month profit
0 1101 8 2000
1 1102 8 4000
2 1103 8 5000
3 1104 8 3500
========行索引========
RangeIndex(start=0, stop=4, step=1)
========列索引========
Index(['apart', 'month', 'profit'], dtype='object')
========数据值========
[['1101' 8 2000]
['1102' 8 4000]
['1103' 8 5000]
['1104' 8 3500]]

2、DataFrame数据获取

import numpy as np
import pandas as pd
from pandas import Series,DataFrame

#3.DataFrame获取数据
data = {
  "apart":['1101',"1102","1103","1104"],
  "profit":[2000,4000,5000,3500],
  "month":8
}

d3 = DataFrame(data)
print(d3)

print("======获取一列数据======")
print(d3["apart"])
print("======获取一行数据======")
print(d3.ix[1])

print("======修改数据值======")
d3["month"] = [7,8,9,10]        #修改值
d3["year"] = [2001,2001,2003,2004]   #新增列
d3.ix["4"] = np.NaN
print(d3)

运行结果：

apart month profit
0 1101 8 2000
1 1102 8 4000
2 1103 8 5000
3 1104 8 3500
======获取一列数据======
0 1101
1 1102
2 1103
3 1104
Name: apart, dtype: object
======获取一行数据======
apart 1102
month 8
profit 4000
Name: 1, dtype: object
======修改数据值======
apart month profit year
0 1101 7.0 2000.0 2001.0
1 1102 8.0 4000.0 2001.0
2 1103 9.0 5000.0 2003.0
3 1104 10.0 3500.0 2004.0
4 NaN NaN NaN NaN

3、pandas基本功能

（1）pandas数据文件读取

import numpy as np
import pandas as pd
from pandas import Series,DataFrame

#pandas基本操作
#1.数据文件读取

df = pd.read_csv("data.csv")
print(df)

运行结果：

name age source
0 gerry 18 98.5
1 tom 21 78.2
2 lili 24 98.5
3 john 20 89.2

（2）数据过滤获取

import numpy as np
import pandas as pd
from pandas import Series,DataFrame

#pandas基本操作
#1.数据文件读取

df = pd.read_csv("data.csv")
print(df)

#2.数据过滤获取

columns = ["姓名","年龄","成绩"]
df.columns = columns    #更改列索引
print("=======更改列索引========")
print(df)

#获取几列的值
df1 = df[columns[1:]]
print("=======获取几列的值========")
print(df1)
print("=======获取几行的值========")
print(df.ix[1:3])

#删除含有NaN值的行
df2 = df1.dropna()
print("=======删除含有NaN值的行=======")
print(df2)

运行结果：

name age source
0 gerry 18 98.5
1 tom 21 NaN
2 lili 24 98.5
3 john 20 89.2
=======更改列索引========
姓名年龄成绩
0 gerry 18 98.5
1 tom 21 NaN
2 lili 24 98.5
3 john 20 89.2
=======获取几列的值========
年龄成绩
0 18 98.5
1 21 NaN
2 24 98.5
3 20 89.2
=======获取几行的值========
姓名年龄成绩
1 tom 21 NaN
2 lili 24 98.5
3 john 20 89.2
=======删除含有NaN值的行=======
年龄成绩
0 18 98.5
2 24 98.5
3 20 89.2