📜  如何在 Pandas 中操作字符串?

📅  最后修改于: 2022-05-13 01:55:43.870000             🧑  作者: Mango

如何在 Pandas 中操作字符串?

Pandas 库提供了多种方法,可用于根据所需的输出操作字符串。但首先,让我们创建一个 Pandas 数据框。

Python3
import pandas as pd
  
data = [[1, "ABC KUMAR", "xYZ"], [2, "BCD", "XXY"],
        [3, "CDE KUMAR", "ZXX"], [3, "DEF", "xYZZ"]]
  
cfile = pd.DataFrame(data, columns = ["SN", "FirstName", "LastName"])
  
cfile


Python3
# find firstname starting with 'D'
result = cfile.FirstName.str.startswith('D')
print(result)
  
# find lasttname containing 'XX'
result = cfile.LastName.str.contains('XX')
print(result)
  
  
# split FirstName on the basis of ' '
result = cfile.FirstName.str.split()
print(result)
  
  
# find length of lasttname
result = cfile.LastName.str.len()
print(result)
  
# Capitalize the first Letter of LastName
result = cfile.LastName.str.capitalize()
print(result)
  
# Capitalize all Letter of LastName
result = cfile.LastName.str.upper()
print(result)
  
# Convert all Letter of LastName to lowercase
result = cfile.LastName.str.lower()
print(result)


输出:

“Pandas”库提供了一个“ .str()”方法,可用于将数据框的任何数据创建为字符串,之后Python文档或本文中定义的任何字符串操作都可以用于该数据。

下面是说明一些示例的代码

Python3

# find firstname starting with 'D'
result = cfile.FirstName.str.startswith('D')
print(result)
  
# find lasttname containing 'XX'
result = cfile.LastName.str.contains('XX')
print(result)
  
  
# split FirstName on the basis of ' '
result = cfile.FirstName.str.split()
print(result)
  
  
# find length of lasttname
result = cfile.LastName.str.len()
print(result)
  
# Capitalize the first Letter of LastName
result = cfile.LastName.str.capitalize()
print(result)
  
# Capitalize all Letter of LastName
result = cfile.LastName.str.upper()
print(result)
  
# Convert all Letter of LastName to lowercase
result = cfile.LastName.str.lower()
print(result)

输出:

0    False
1    False
2    False
3     True
Name: FirstName, dtype: bool
0    False
1     True
2     True
3    False
Name: LastName, dtype: bool
0    [ABC, KUMAR]
1           [BCD]
2    [CDE, KUMAR]
3           [DEF]
Name: FirstName, dtype: object
0    3
1    3
2    3
3    4
Name: LastName, dtype: int64
0     Xyz
1     Xxy
2     Zxx
3    Xyzz
Name: LastName, dtype: object
0     XYZ
1     XXY
2     ZXX
3    XYZZ
Name: LastName, dtype: object
0     xyz
1     xxy
2     zxx
3    xyzz
Name: LastName, dtype: object