pandas的数据操作

1.数据类型映射

2.过滤操作

sql中的where语句的功能非常丰富，常用关键包括 =，<>，>=，<=，>，<，in，not in，isnull，like，and，or等关键字，下面我们就来看看，如果是在pandas中该如何实现。

>>> import pandas as pd
>>> import numpy as np
df = pd.read_excel(r'D:/myExcel/1.xlsx')
>>> df = pd.read_excel(r'D:/myExcel/1.xlsx')
>>> df
  id   name  score grade
0  a    bog   45.0     A
1  c  jiken   67.0     B
2  d    bob   23.0     A
3  b  jiken   34.0     B
4  f   lucy    NaN     A
5  e   tidy   75.0     B
1、==
即判断相等关系，和sql中的=类似

# 获取id中为'a'的行
>>> df[df['id'] =='a']
  id name  score grade
0  a  bog   45.0     A
# 获取score中分数为45的行
>>> df[df['score'] == 45]
  id name  score grade
0  a  bog   45.0     A
2、!=
即表明不相等的关系，和sql中的<>类似

# 获取id中不为'a'的行
>>> df[df['id'] !='a']
  id   name  score grade
1  c  jiken   67.0     B
2  d    bob   23.0     A
3  b  jiken   34.0     B
4  f   lucy    NaN     A
5  e   tidy   75.0     B
# 获取score中不等于45的行
>>> df[df['score'] != 45]
  id   name  score grade
1  c  jiken   67.0     B
2  d    bob   23.0     A
3  b  jiken   34.0     B
4  f   lucy    NaN     A
5  e   tidy   75.0     B
3、>,>=,<,<=
此和sql保持一致

# score需>= 45
>>> df[df['score'] >= 45]
  id   name  score grade
0  a    bog   45.0     A
1  c  jiken   67.0     B
5  e   tidy   75.0     B
# score需> 45
>>> df[df['score'] > 45]
  id   name  score grade
1  c  jiken   67.0     B
5  e   tidy   75.0     B
# score小于45
>>> df[df['score'] < 45]
  id   name  score grade
2  d    bob   23.0     A
3  b  jiken   34.0     B
# score小于等于45
>>> df[df['score'] <= 45]
  id   name  score grade
0  a    bog   45.0     A
2  d    bob   23.0     A
3  b  jiken   34.0     B

2.替换操作

# 假如表中有88，则88全部替换为99
df.replace({88:99})

3.替换df中某一行的特定值

# index代表行索引，2代表列的索引值为2
df.iloc[index, 2] = ""