精选23个Pandas函数

作者: 皮皮大 | 来源:发表于2022-01-14 10:31 被阅读0次

公众号：尤而小屋
作者：Peter
编辑：Peter

大家好，我是Peter~

从26个字母中精选出23个Pandas常用的函数，将它们的使用方法介绍给大家。其中o、y、z没有相应的函数。

image

import pandas as pd
import numpy as np

下面介绍每个函数的使用方法，更多详细的内容请移步官网：https://pandas.pydata.org/docs/reference/general_functions.html

assign函数

df = pd.DataFrame({
    'temp_c': [17.0, 25.0]},
    index=['Portland', 'Berkeley'])
df

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>temp_c</th>
</tr>
</thead>
<tbody>
<tr>
<th>Portland</th>
<td>17.0</td>
</tr>
<tr>
<th>Berkeley</th>
<td>25.0</td>
</tr>
</tbody>
</table>

</div>

# 生成新的字段

df.assign(temp_f=df['temp_c'] * 9 / 5 + 32)

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>temp_c</th>
<th>temp_f</th>
</tr>
</thead>
<tbody>
<tr>
<th>Portland</th>
<td>17.0</td>
<td>62.6</td>
</tr>
<tr>
<th>Berkeley</th>
<td>25.0</td>
<td>77.0</td>
</tr>
</tbody>
</table>

</div>

df  # 原来DataFrame是不改变的

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>temp_c</th>
</tr>
</thead>
<tbody>
<tr>
<th>Portland</th>
<td>17.0</td>
</tr>
<tr>
<th>Berkeley</th>
<td>25.0</td>
</tr>
</tbody>
</table>

</div>

df["temp_f1"] = df["temp_c"] * 9 / 5 + 32
df

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>temp_c</th>
<th>temp_f1</th>
</tr>
</thead>
<tbody>
<tr>
<th>Portland</th>
<td>17.0</td>
<td>62.6</td>
</tr>
<tr>
<th>Berkeley</th>
<td>25.0</td>
<td>77.0</td>
</tr>
</tbody>
</table>

</div>

df

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>temp_c</th>
<th>temp_f1</th>
</tr>
</thead>
<tbody>
<tr>
<th>Portland</th>
<td>17.0</td>
<td>62.6</td>
</tr>
<tr>
<th>Berkeley</th>
<td>25.0</td>
<td>77.0</td>
</tr>
</tbody>
</table>

</div>

bool函数

返回单个Series或者DataFrame中单个元素的bool值：True或者False

pd.Series([True]).bool()

True

pd.Series([False]).bool()

False

pd.DataFrame({'col': [True]}).bool()

True

pd.DataFrame({'col': [False]}).bool()

False

# # 多个元素引发报错

# pd.DataFrame({'col': [True,False]}).bool()

image

concat函数

该函数是用来表示多个DataFrame的拼接，横向或者纵向皆可。

df1 = pd.DataFrame({
    "sid":["s1","s2"],
    "name":["xiaoming","Mike"]})
df1

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>xiaoming</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>Mike</td>
</tr>
</tbody>
</table>

</div>

df2 = pd.DataFrame({
    "sid":["s3","s4"],
    "name":["Tom","Peter"]})
df2

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s3</td>
<td>Tom</td>
</tr>
<tr>
<th>1</th>
<td>s4</td>
<td>Peter</td>
</tr>
</tbody>
</table>

</div>

df3 = pd.DataFrame({
    "address":["北京","深圳"],             
    "sex":["Male","Female"]})
df3

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>address</th>
<th>sex</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>北京</td>
<td>Male</td>
</tr>
<tr>
<th>1</th>
<td>深圳</td>
<td>Female</td>
</tr>
</tbody>
</table>

</div>

# 使用1：纵向
pd.concat([df1,df2])

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>xiaoming</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>Mike</td>
</tr>
<tr>
<th>0</th>
<td>s3</td>
<td>Tom</td>
</tr>
<tr>
<th>1</th>
<td>s4</td>
<td>Peter</td>
</tr>
</tbody>
</table>

</div>

# 使用2：横向
pd.concat([df1,df3],axis=1)

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
<th>address</th>
<th>sex</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>xiaoming</td>
<td>北京</td>
<td>Male</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>Mike</td>
<td>深圳</td>
<td>Female</td>
</tr>
</tbody>
</table>

</div>

dropna函数

删除空值

df4 = pd.DataFrame({
    "sid":["s1","s2", np.nan],             
    "name":["xiaoming",np.nan, "Mike"]})
df4

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>xiaoming</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>NaN</td>
</tr>
<tr>
<th>2</th>
<td>NaN</td>
<td>Mike</td>
</tr>
</tbody>
</table>

</div>

df4.dropna()

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>xiaoming</td>
</tr>
</tbody>
</table>

</div>

df4.dropna(subset=["name"])

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>xiaoming</td>
</tr>
<tr>
<th>2</th>
<td>NaN</td>
<td>Mike</td>
</tr>
</tbody>
</table>

</div>

explode函数

爆炸函数的使用：将宽表转成长表

df5 = pd.DataFrame({
    "sid":["s1","s2"],       
    "phones":[["华为","小米","一加"],["三星","苹果"]]
                   })
df5

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>phones</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>[华为, 小米, 一加]</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>[三星, 苹果]</td>
</tr>
</tbody>
</table>

</div>

df5.explode("phones")

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>phones</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>华为</td>
</tr>
<tr>
<th>0</th>
<td>s1</td>
<td>小米</td>
</tr>
<tr>
<th>0</th>
<td>s1</td>
<td>一加</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>三星</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>苹果</td>
</tr>
</tbody>
</table>

</div>

df5

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>phones</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>[华为, 小米, 一加]</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>[三星, 苹果]</td>
</tr>
</tbody>
</table>

</div>

fillna函数

填充缺失值

df4

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>xiaoming</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>NaN</td>
</tr>
<tr>
<th>2</th>
<td>NaN</td>
<td>Mike</td>
</tr>
</tbody>
</table>

</div>

df4.fillna({"sid":"s3","name":"Peter"})

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>xiaoming</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>Peter</td>
</tr>
<tr>
<th>2</th>
<td>s3</td>
<td>Mike</td>
</tr>
</tbody>
</table>

</div>

groupby函数

同组统计的功能

# 借用这个结果
df6 = df5.explode("phones")
df6

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>phones</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>华为</td>
</tr>
<tr>
<th>0</th>
<td>s1</td>
<td>小米</td>
</tr>
<tr>
<th>0</th>
<td>s1</td>
<td>一加</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>三星</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>苹果</td>
</tr>
</tbody>
</table>

</div>

df6.groupby("sid")["phones"].count()

sid
s1    3
s2    2
Name: phones, dtype: int64

head函数

查看前几行的数据，默认是前5行

df7 = pd.DataFrame({
    "sid":list(range(10)),                
    "name":list(range(80,100,2))})
df7

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

df7.head()   # 默认前5行

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

df7.head(3)  # 指定前3行

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

isnull函数

判断是否存在缺失值，超级常用的函数

df4

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>xiaoming</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>NaN</td>
</tr>
<tr>
<th>2</th>
<td>NaN</td>
<td>Mike</td>
</tr>
</tbody>
</table>

</div>

df4.isnull()  # True表示缺失

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>name</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>False</td>
<td>False</td>
</tr>
<tr>
<th>1</th>
<td>False</td>
<td>True</td>
</tr>
<tr>
<th>2</th>
<td>True</td>
<td>False</td>
</tr>
</tbody>
</table>

</div>

df4.isnull().sum()  # 每个字段缺失的总和

sid     1
name    1
dtype: int64

df6.isnull().sum()   # 没有缺失值

sid       0
phones    0
dtype: int64

join函数

用于连接不同的DataFrame：

df7 = pd.DataFrame({
    'key': ['K0', 'K1', 'K2', 'K3', 'K4', 'K5'],
    'A': ['A0', 'A1', 'A2', 'A3', 'A4', 'A5']})
df7

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

df8 = pd.DataFrame({
    'key': ['K0', 'K1', 'K2'],
    'B': ['B0', 'B1', 'B2']})
df8

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

df7.join(df8,lsuffix="_df7",rsuffix="_df8")

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

kurt函数

查找数据的峰度值

df9 = pd.DataFrame({
    "A":[12, 4, 5, 44, 1], 
    "B":[5, 2, 54, 3, 2], 
    "C":[20, 16, 7, 3, 8], 
    "D":[14, 3, 17, 2, 6]}) 
df9

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

df9.kurt()

A    3.936824
B    4.941512
C   -1.745717
D   -2.508808
dtype: float64

loc函数

loc就是location的缩写，定位查找数据

df9

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

df9.loc[1,:]  # 第一行全部列的数据

A     4
B     2
C    16
D     3
Name: 1, dtype: int64

df9.loc[1:3,"B"]  # 1到3行的B列

1     2
2    54
3     3
Name: B, dtype: int64

merge函数

同样也是数据的合并函数，类似SQL中的join，功能最为强大

df7

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

df8

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

pd.merge(df7,df8)  # 默认how的参数是inner

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

pd.merge(df7,df8,how="outer")

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

nunique函数

用于统计数据的唯一值

df10 = pd.DataFrame({
    "sid":list("acbdefg"),
    "score":[9,8,9,7,8,9,3]
                    })
df10

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>score</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>a</td>
<td>9</td>
</tr>
<tr>
<th>1</th>
<td>c</td>
<td>8</td>
</tr>
<tr>
<th>2</th>
<td>b</td>
<td>9</td>
</tr>
<tr>
<th>3</th>
<td>d</td>
<td>7</td>
</tr>
<tr>
<th>4</th>
<td>e</td>
<td>8</td>
</tr>
<tr>
<th>5</th>
<td>f</td>
<td>9</td>
</tr>
<tr>
<th>6</th>
<td>g</td>
<td>3</td>
</tr>
</tbody>
</table>

</div>

df10.nunique()

sid      7
score    4
dtype: int64

pct_change函数

计算当前时期和前一个时期的比值

s = pd.Series([90, 91, 85])
s

0    90
1    91
2    85
dtype: int64

s.pct_change()

0         NaN
1    0.011111
2   -0.065934
dtype: float64

(91 - 90) / 90

0.011111111111111112

(85 - 91) / 91

-0.06593406593406594

# 和前两个时期相比
s.pct_change(periods=2)

0         NaN
1         NaN
2   -0.055556
dtype: float64

# 如果存在空值，用填充方法
s = pd.Series([90, 91, None, 85])
s

0    90.0
1    91.0
2     NaN
3    85.0
dtype: float64

s.pct_change(fill_method='ffill')

0         NaN
1    0.011111
2    0.000000
3   -0.065934
dtype: float64

query函数

根据条件查询取值

df10

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>score</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>a</td>
<td>9</td>
</tr>
<tr>
<th>1</th>
<td>c</td>
<td>8</td>
</tr>
<tr>
<th>2</th>
<td>b</td>
<td>9</td>
</tr>
<tr>
<th>3</th>
<td>d</td>
<td>7</td>
</tr>
<tr>
<th>4</th>
<td>e</td>
<td>8</td>
</tr>
<tr>
<th>5</th>
<td>f</td>
<td>9</td>
</tr>
<tr>
<th>6</th>
<td>g</td>
<td>3</td>
</tr>
</tbody>
</table>

</div>

df10.query("score >= 8")

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>score</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>a</td>
<td>9</td>
</tr>
<tr>
<th>1</th>
<td>c</td>
<td>8</td>
</tr>
<tr>
<th>2</th>
<td>b</td>
<td>9</td>
</tr>
<tr>
<th>4</th>
<td>e</td>
<td>8</td>
</tr>
<tr>
<th>5</th>
<td>f</td>
<td>9</td>
</tr>
</tbody>
</table>

</div>

rank函数

进行排名的函数，类似SQL的窗口函数功能：

df10

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>score</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>a</td>
<td>9</td>
</tr>
<tr>
<th>1</th>
<td>c</td>
<td>8</td>
</tr>
<tr>
<th>2</th>
<td>b</td>
<td>9</td>
</tr>
<tr>
<th>3</th>
<td>d</td>
<td>7</td>
</tr>
<tr>
<th>4</th>
<td>e</td>
<td>8</td>
</tr>
<tr>
<th>5</th>
<td>f</td>
<td>9</td>
</tr>
<tr>
<th>6</th>
<td>g</td>
<td>3</td>
</tr>
</tbody>
</table>

</div>

df10["rank_10"] = df10["score"].rank()
df10

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>score</th>
<th>rank_10</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>a</td>
<td>9</td>
<td>6.0</td>
</tr>
<tr>
<th>1</th>
<td>c</td>
<td>8</td>
<td>3.5</td>
</tr>
<tr>
<th>2</th>
<td>b</td>
<td>9</td>
<td>6.0</td>
</tr>
<tr>
<th>3</th>
<td>d</td>
<td>7</td>
<td>2.0</td>
</tr>
<tr>
<th>4</th>
<td>e</td>
<td>8</td>
<td>3.5</td>
</tr>
<tr>
<th>5</th>
<td>f</td>
<td>9</td>
<td>6.0</td>
</tr>
<tr>
<th>6</th>
<td>g</td>
<td>3</td>
<td>1.0</td>
</tr>
</tbody>
</table>

</div>

df10["rank_10_max"] = df10["score"].rank(method="max")
df10

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>score</th>
<th>rank_10</th>
<th>rank_10_max</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>a</td>
<td>9</td>
<td>6.0</td>
<td>7.0</td>
</tr>
<tr>
<th>1</th>
<td>c</td>
<td>8</td>
<td>3.5</td>
<td>4.0</td>
</tr>
<tr>
<th>2</th>
<td>b</td>
<td>9</td>
<td>6.0</td>
<td>7.0</td>
</tr>
<tr>
<th>3</th>
<td>d</td>
<td>7</td>
<td>2.0</td>
<td>2.0</td>
</tr>
<tr>
<th>4</th>
<td>e</td>
<td>8</td>
<td>3.5</td>
<td>4.0</td>
</tr>
<tr>
<th>5</th>
<td>f</td>
<td>9</td>
<td>6.0</td>
<td>7.0</td>
</tr>
<tr>
<th>6</th>
<td>g</td>
<td>3</td>
<td>1.0</td>
<td>1.0</td>
</tr>
</tbody>
</table>

</div>

df10["rank_10_min"] = df10["score"].rank(method="min")
df10

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>score</th>
<th>rank_10</th>
<th>rank_10_max</th>
<th>rank_10_min</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>a</td>
<td>9</td>
<td>6.0</td>
<td>7.0</td>
<td>5.0</td>
</tr>
<tr>
<th>1</th>
<td>c</td>
<td>8</td>
<td>3.5</td>
<td>4.0</td>
<td>3.0</td>
</tr>
<tr>
<th>2</th>
<td>b</td>
<td>9</td>
<td>6.0</td>
<td>7.0</td>
<td>5.0</td>
</tr>
<tr>
<th>3</th>
<td>d</td>
<td>7</td>
<td>2.0</td>
<td>2.0</td>
<td>2.0</td>
</tr>
<tr>
<th>4</th>
<td>e</td>
<td>8</td>
<td>3.5</td>
<td>4.0</td>
<td>3.0</td>
</tr>
<tr>
<th>5</th>
<td>f</td>
<td>9</td>
<td>6.0</td>
<td>7.0</td>
<td>5.0</td>
</tr>
<tr>
<th>6</th>
<td>g</td>
<td>3</td>
<td>1.0</td>
<td>1.0</td>
<td>1.0</td>
</tr>
</tbody>
</table>

</div>

sort_values函数

根据数据进行排序的函数

df9

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

df9.sort_values("A")  # 默认是升序排列

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

# 先根据B升序，如果B相同，再根据D降序

df9.sort_values(["B","D"], ascending=[True,False])

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

tail函数

查看末尾的数据

df7.tail()

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

df7.tail(3)

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

</div>

unique函数

查找每个字段的唯一元素

df10

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>score</th>
<th>rank_10</th>
<th>rank_10_max</th>
<th>rank_10_min</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>a</td>
<td>9</td>
<td>6.0</td>
<td>7.0</td>
<td>5.0</td>
</tr>
<tr>
<th>1</th>
<td>c</td>
<td>8</td>
<td>3.5</td>
<td>4.0</td>
<td>3.0</td>
</tr>
<tr>
<th>2</th>
<td>b</td>
<td>9</td>
<td>6.0</td>
<td>7.0</td>
<td>5.0</td>
</tr>
<tr>
<th>3</th>
<td>d</td>
<td>7</td>
<td>2.0</td>
<td>2.0</td>
<td>2.0</td>
</tr>
<tr>
<th>4</th>
<td>e</td>
<td>8</td>
<td>3.5</td>
<td>4.0</td>
<td>3.0</td>
</tr>
<tr>
<th>5</th>
<td>f</td>
<td>9</td>
<td>6.0</td>
<td>7.0</td>
<td>5.0</td>
</tr>
<tr>
<th>6</th>
<td>g</td>
<td>3</td>
<td>1.0</td>
<td>1.0</td>
<td>1.0</td>
</tr>
</tbody>
</table>

</div>

df10["score"].unique()

array([9, 8, 7, 3])

df10["rank_10"].unique()

array([6. , 3.5, 2. , 1. ])

value_counts函数

用于统计字段中每个唯一值的个数

df6

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>sid</th>
<th>phones</th>
</tr>
</thead>
<tbody>
<tr>
<th>0</th>
<td>s1</td>
<td>华为</td>
</tr>
<tr>
<th>0</th>
<td>s1</td>
<td>小米</td>
</tr>
<tr>
<th>0</th>
<td>s1</td>
<td>一加</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>三星</td>
</tr>
<tr>
<th>1</th>
<td>s2</td>
<td>苹果</td>
</tr>
</tbody>
</table>

</div>

df6["sid"].value_counts()

s1    3
s2    2
Name: sid, dtype: int64

df6["phones"].value_counts()

华为    1
苹果    1
三星    1
一加    1
小米    1
Name: phones, dtype: int64

where函数

用于查找Series或者DataFrame中满足某个条件的数据

w = pd.Series(range(7))
w

0    0
1    1
2    2
3    3
4    4
5    5
6    6
dtype: int64

# 满足条件的显示；不满足的用空值代替
w.where(w>3)

0    NaN
1    NaN
2    NaN
3    NaN
4    4.0
5    5.0
6    6.0
dtype: float64

# 不满足条件的用8代替
w.where(w > 1, 8)

0    8
1    8
2    2
3    3
4    4
5    5
6    6
dtype: int64

xs函数

该函数是用于多层级索引中用于获取指定索引处的值，使用一个关键参数来选择多索引特定级别的数据。

d = {'num_legs': [4, 4, 2, 2],
     'num_wings': [0, 0, 2, 2],
     'class': ['mammal', 'mammal', 'mammal', 'bird'],
     'animal': ['cat', 'dog', 'bat', 'penguin'],
     'locomotion': ['walks', 'walks', 'flies', 'walks']}
# 生成数据
df11 = pd.DataFrame(data=d)
# 重置索引
df11 = df11.set_index(['class', 'animal', 'locomotion'])
df11

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th></th>
<th></th>
<th>num_legs</th>
<th>num_wings</th>
</tr>
<tr>
<th>class</th>
<th>animal</th>
<th>locomotion</th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<th rowspan="3" valign="top">mammal</th>
<th>cat</th>
<th>walks</th>
<td>4</td>
<td>0</td>
</tr>
<tr>
<th>dog</th>
<th>walks</th>
<td>4</td>
<td>0</td>
</tr>
<tr>
<th>bat</th>
<th>flies</th>
<td>2</td>
<td>2</td>
</tr>
<tr>
<th>bird</th>
<th>penguin</th>
<th>walks</th>
<td>2</td>
<td>2</td>
</tr>
</tbody>
</table>

</div>

# 获取指定索引的值
df11.xs('mammal')

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th></th>
<th>num_legs</th>
<th>num_wings</th>
</tr>
<tr>
<th>animal</th>
<th>locomotion</th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<th>cat</th>
<th>walks</th>
<td>4</td>
<td>0</td>
</tr>
<tr>
<th>dog</th>
<th>walks</th>
<td>4</td>
<td>0</td>
</tr>
<tr>
<th>bat</th>
<th>flies</th>
<td>2</td>
<td>2</td>
</tr>
</tbody>
</table>

</div>

# 指定多个索引处的值
df11.xs(('mammal', 'dog'))

/Applications/downloads/anaconda/anaconda3/lib/python3.7/site-packages/IPython/core/interactiveshell.py:2881: PerformanceWarning: indexing past lexsort depth may impact performance.
  return runner(coro)

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>num_legs</th>
<th>num_wings</th>
</tr>
<tr>
<th>locomotion</th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<th>walks</th>
<td>4</td>
<td>0</td>
</tr>
</tbody>
</table>

</div>

# 获取指定索引和级别(level)的值

df11.xs('cat', level=1)

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th></th>
<th>num_legs</th>
<th>num_wings</th>
</tr>
<tr>
<th>class</th>
<th>locomotion</th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<th>mammal</th>
<th>walks</th>
<td>4</td>
<td>0</td>
</tr>
</tbody>
</table>

</div>

df11

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th></th>
<th></th>
<th>num_legs</th>
<th>num_wings</th>
</tr>
<tr>
<th>class</th>
<th>animal</th>
<th>locomotion</th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<th rowspan="3" valign="top">mammal</th>
<th>cat</th>
<th>walks</th>
<td>4</td>
<td>0</td>
</tr>
<tr>
<th>dog</th>
<th>walks</th>
<td>4</td>
<td>0</td>
</tr>
<tr>
<th>bat</th>
<th>flies</th>
<td>2</td>
<td>2</td>
</tr>
<tr>
<th>bird</th>
<th>penguin</th>
<th>walks</th>
<td>2</td>
<td>2</td>
</tr>
</tbody>
</table>

</div>

# 获取多个索引和级别的值
df11.xs(('bird', 'walks'),level=[0, 'locomotion'])

<div>
<style scoped>
.dataframe tbody tr th:only-of-type {
vertical-align: middle;
}

.dataframe tbody tr th {
    vertical-align: top;
}

.dataframe thead th {
    text-align: right;
}

</style>

<table border="1" class="dataframe">
<thead>
<tr style="text-align: right;">
<th></th>
<th>num_legs</th>
<th>num_wings</th>
</tr>
<tr>
<th>animal</th>
<th></th>
<th></th>
</tr>
</thead>
<tbody>
<tr>
<th>penguin</th>
<td>2</td>
<td>2</td>
</tr>
</tbody>
</table>

</div>

# 获取指定列和轴上的值
df11.xs('num_wings', axis=1)

class   animal   locomotion
mammal  cat      walks         0
        dog      walks         0
        bat      flies         2
bird    penguin  walks         2
Name: num_wings, dtype: int64

网友评论

本文标题：精选23个Pandas函数

本文链接：https://www.haomeiwen.com/subject/lpydcrtx.html

延伸阅读

深度阅读

您也可以注册成为美文阅读网的作者，发表您的原创作品、分享您的心情！

精选23个Pandas函数

assign函数

bool函数

concat函数

dropna函数

explode函数

fillna函数

groupby函数

head函数

isnull函数

join函数

kurt函数

loc函数

merge函数

nunique函数

pct_change函数

query函数

rank函数

sort_values函数

tail函数

unique函数

value_counts函数

where函数

xs函数

相关文章

网友评论

延伸阅读

深度阅读

栏目导航

热点阅读