Pandas基础笔记

1、安装Pandas

常用的方法是通过Anaconda安装，在终端或命令符输入如下命令安装：

conda install pandas
若未安装Anaconda，使用Python自带的包管理工具pip来安装：

pip install pandas

2、导入pandas库和查询相应的版本信息

import numpy as np   # pandas和numpy常常结合在一起使用，导入numpy库
import pandas as pd  # 导入pandas库
print(pd.
__version__
) # 打印pandas版本信息
#> 0.23.4

3、pandas数据类型

pandas包含两种数据类型：series和dataframe。

series是一种一维数据结构，每一个元素都带有一个索引，与一维数组的含义相似，其中索引可以为数字或字符串。series结构名称：

dataframe是一种二维数据结构，数据以表格形式（与excel类似）存储，有对应的行和列。dataframe结构名称：

4、series教程

从列表，数组，字典构建series

mylist = list('abcedfghijklmnopqrstuvwxyz')   # 列表
myarr = np.arange(26)                       # 数组
mydict = dict(zip(mylist, myarr))             # 字典
构建方法
ser1 = pd.Series(mylist)
ser2 = pd.Series(myarr)
ser3 = pd.Series(mydict)
print(ser3.head())                 # 打印前5个数据
#>  a    0
    b    1
    c    2
    d    4
    e    3
    dtype:int64

使series的索引列转化为dataframe的列

mylist = list('abcedfghijklmnopqrstuvwxyz')
myarr = np.arange(26)
mydict = dict(zip(mylist, myarr))
ser = pd.Series(mydict)
series转换为dataframe
df = ser.to_frame()
索引列转换为dataframe的列
df.reset_index(inplace=True)
print(df.head())
#>   index  0
0     a  0
1     b  1
2     c  2
3     e  3
4     d  4

结合多个series组成dataframe

构建series1
ser1 = pd.Series(list('abcedfghijklmnopqrstuvwxyz')) 
构建series2
ser2 = pd.Series(np.arange(26))
方法1，axis=1表示列拼接，0表示行拼接
df = pd.concat([ser1, ser2], axis=1)
与方法1相比，方法2设置了列名
df = pd.DataFrame({'col1': ser1, 'col2': ser2})
print(df.head())
#>   col1  col2
0    a     0
1    b     1
2    c     2
3    e     3
4    d     4

命名列索引的名称

ser = pd.Series(list('abcedfghijklmnopqrstuvwxyz'))
命名索引列名称
ser.name = 'alphabets'
显示前5行数据
ser.head()
#>        0    a
1    b
2    c
3    e
4    d
 Name: alphabets, dtype: object

获得series对象A中不包含series对象B的元素

ser1 = pd.Series([1, 2, 3, 4, 5])
ser2 = pd.Series([4, 5, 6, 7, 8])
返回ser1不包含ser2的布尔型series
ser3=~ser1.isin(ser2)
获取ser不包含ser2的元素
ser1[ser3]
#>        0    1
1    2
2    3
 dtype: int64

获得seriesA和seriesB不相同的项

ser1 = pd.Series([1, 2, 3, 4, 5])
ser2 = pd.Series([4, 5, 6, 7, 8])
求ser1和ser2的并集
ser_u = pd.Series(np.union1d(ser1, ser2))
求ser1和ser2的交集
ser_i = pd.Series(np.intersect1d(ser1, ser2))
ser_i在ser_u的补集就是ser1和ser2不相同的项
ser_u[~ser_u.isin(ser_i)]
#>        0    1
1    2
2    3
5    6
6    7
7    8
 dtype: int64

获得数值series的四分位值

设置随机数种子
state = np.random.RandomState(100)
从均值为5标准差为25的正态分布随机抽取5个点构成series
ser = pd.Series(state.normal(10, 5, 25))
求ser的四分位数
np.percentile(ser, q=[0, 25, 50, 75, 100])
#> array([ 1.25117263,  7.70986507, 10.92259345, 13.36360403, 18.0949083 ])

获得series中单一项的频率计数

#从0~7随机抽取30个列表值，组成series
ser = pd.Series(np.take(list('abcdefgh'), np.random.randint(8, size=30)))
对该series进行计数
ser.value_counts()
#>        d    8
 g    6
 b    6
 a    5
 e    2
 h    2
 f    1
 dtype: int64

保留series中前两个频次最多的项，其他项替换为‘other’

np.random.RandomState(100)
从1~4均匀采样12个点组成series
ser = pd.Series(np.random.randint(1, 5, [12]))
除前两行索引对应的值不变，后几行索引对应的值为Other
ser[~ser.isin(ser.value_counts().index[:2])] = 'Other'
ser
#>        0     Other
1         4
2         2
3         2
4         4
5     Other
6     Other
7     Other
8         4
9         4
10        4
11        2
 dtype: object

对数值series分成10个相同数目的组
换个角度理解，对数值series离散化成10个类别（categorical）值

ser = pd.Series(np.random.random(20))
离散化10个类别值,只显示前5行的数据
pd.qcut(ser, q=[0, .10, .20, .3, .4, .5, .6, .7, .8, .9, 1],
        labels=['1st', '2nd', '3rd', '4th', '5th', '6th', '7th', '8th', '9th', '10th']).head()
#>
0    3rd
1    1st
2    6th
3    6th
4    9th
dtype: category
Categories (10, object): [1st < 2nd < 3rd < 4th ... 7th < 8th < 9th < 10th]

使numpy数组转化为给定形状的dataframe

ser = pd.Series(np.random.randint(1, 10, 35))
serier类型转换numpy类型，然后重构
df = pd.DataFrame(ser.values.reshape(7,5))
print(df)
#>    0  1  2  3  4
0  1  2  1  2  5
1  1  2  4  5  2
2  1  3  3  2  8
3  8  6  4  9  6
4  2  1  1  8  5
5  3  2  8  5  6
6  1  5  5  4  6

找到series的值是3的倍数的位置

ser = pd.Series(np.random.randint(1, 10, 7))
print(ser)
获取值是3倍数的索引
np.argwhere(ser % 3==0)
#>        0    6
1    8
2    6
3    7
4    6
5    2
6    4
 dtype: int64
 
#>        array([[0],
        [2],
        [4]])

获取series中给定索引的元素（items）

ser = pd.Series(list('abcdefghijklmnopqrstuvwxyz'))
index = [0, 4, 8, 14, 20]
获取指定索引的元素
ser.take(index)
#>        0     a
4     e
8     i
14    o
20    u
 dtype: object

垂直和水平的拼接series

ser1 = pd.Series(range(5))
ser2 = pd.Series(list('abcde'))
垂直拼接
df = pd.concat([ser1, ser2], axis=0)
水平拼接
df = pd.concat([ser1, ser2], axis=1)
print(df)
#>    0  1
0  0  a
1  1  b
2  2  c
3  3  d
4  4  e

获取series对象A中包含series对象B元素的位置

ser1必须包含ser2，否则会报错
ser1 = pd.Series([10, 9, 6, 5, 3, 1, 12, 8, 13])
ser2 = pd.Series([1, 3, 10, 13])
方法 1
[np.where(i == ser1)[0].tolist()[0] for i in ser2]
方法 2
[pd.Index(ser1).get_loc(i) for i in ser2]
#> [5, 4, 0, 8]

计算series之间的均方差

truth = pd.Series(range(10))
pred = pd.Series(range(10)) + np.random.random(10)
均方差
np.mean((truth-pred)**2)
#> 0.25508722434194103

使series中每个元素的首字母为大写

series的元素为str类型
ser = pd.Series(['how', 'to', 'kick', 'ass?'])
方法 1
ser.map(lambda x: x.title())
方法 2 ，字符串相加
ser.map(lambda x: x[0].upper() + x[1:])
方法 3
pd.Series([i.title() for i in ser])
#>        0     How
1      To
2    Kick
3    Ass?
 dtype: object

计算series中每个元素的字符串长度

ser = pd.Series(['how', 'to', 'kick', 'ass?'])
方法
ser.map(lambda x: len(x))
#>        0    3
1    2
2    4
3    4
 dtype: int64

计算series的一阶导和二阶导

ser = pd.Series([1, 3, 6, 10, 15, 21, 27, 35])
求一阶导并转化为列表类型
print(ser.diff().tolist())
求二阶导并转化为列表类型
print(ser.diff().diff().tolist())
#>        [nan, 2.0, 3.0, 4.0, 5.0, 6.0, 6.0, 8.0]
 [nan, nan, 1.0, 1.0, 1.0, 1.0, 0.0, 2.0]

将一系列日期字符串转换为timeseries

ser = pd.Series(['01 Jan 2010', '02-02-2011', '20120303', '2013/04/04', '2014-05-05', '2015-06-06T12:20'])
pd.to_datetime(ser)
#>        0   2010-01-01 00:00:00
1   2011-02-02 00:00:00
2   2012-03-03 00:00:00
3   2013-04-04 00:00:00
4   2014-05-05 00:00:00
5   2015-06-06 12:20:00
 dtype: datetime64[ns]

从一个series中获取至少包含两个元音的元素

ser = pd.Series(['Apple', 'Orange', 'Plan', 'Python', 'Money'])
方法
from collections import Counter
Counter是一个类字典类型，键是元素值，值是元素出现的次数,满足条件的元素返回True
mask = ser.map(lambda x: sum([Counter(x.lower()).get(i, 0) for i in list('aeiou')]) >= 2)
ser[mask]
#>        0     Apple
1    Orange
4     Money
 dtype: object

计算根据另一个series分组后的series均值

fruit = pd.Series(np.random.choice(['apple', 'banana', 'carrot'], 10))
weights = pd.Series(np.linspace(1, 10, 10))
根据fruit对weight分组
weightsGrouped = weights.groupby(fruit)
print(weightsGrouped.indices)
对分组后series求每个索引的平均值
weightsGrouped.mean()
#>        {'apple': array([0, 3], dtype=int64), 'banana': array([1, 2, 4, 8],
 dtype=int64), 'carrot': array([5, 6, 7, 9], dtype=int64)}
#>        apple     2.50
 banana    4.75
 carrot    7.75
 dtype: float64

计算两个series之间的欧氏距离

p = pd.Series([1, 2, 3, 4, 5, 6, 7, 8, 9, 10])
q = pd.Series([10, 9, 8, 7, 6, 5, 4, 3, 2, 1])
方法1
sum((p - q)
**2)**
.5
方法2
np.linalg.norm(p-q)
#>        18.16590212458495

在数值series中找局部最大值
局部最大值对应二阶导局部最小值

ser = pd.Series([2, 10, 3, 4, 9, 10, 2, 7, 3])
二阶导
dd = np.diff(np.sign(np.diff(ser)))
二阶导的最小值对应的值为最大值，返回最大值的索引
peak_locs = np.where(dd == -2)[0] + 1
peak_locs
#>        array([1, 5, 7], dtype=int64)

用最少出现的字符替换空格符

my_str = 'dbc deb abed gade'
方法
ser = pd.Series(list('dbc deb abed gade'))
统计元素的频数
freq = ser.value_counts()
print(freq)
求最小频数的字符
least_freq = freq.dropna().index[-1]
替换
"".join(ser.replace(' ', least_freq))
#>        d    4
      3
 b    3
 e    3
 a    2
 c    1
 g    1
 dtype: int64
#>        'dbcgdebgabedggade'

计算数值series的自相关系数

ser = pd.Series(np.arange(20) + np.random.normal(1, 10, 20))
求series的自相关系数，i为偏移量
autocorrelations = [ser.autocorr(i).round(2) for i in range(11)]
print(autocorrelations[1:])
选择最大的偏移量
print('Lag having highest correlation: ', np.argmax(np.abs(autocorrelations[1:]))+1)
#>        [0.33, 0.41, 0.48, 0.01, 0.21, 0.16, -0.11, 0.05, 0.34, -0.24]
#>        Lag having highest correlation:  3

对series进行算术运算操作

如何对series之间进行算法运算
import pandas as pd
series1 = pd.Series([3,4,4,4],['index1','index2','index3','index4'])
series2 = pd.Series([2,2,2,2],['index1','index2','index33','index44'])
加法
series_add = series1 + series2
print(series_add)
减法
series_minus = series1 - series2
series_minus
乘法
series_multi = series1 * series2
series_multi
除法
series_div = series1/series2
series_div
series是基于索引进行算数运算操作的，pandas会根据索引对数据进行运算，若series之间有不同的索引，对应的值就为Nan。结果如下：
#加法：
 index1     5.0
 index2     6.0
 index3     NaN
 index33    NaN
 index4     NaN
 index44    NaN
 dtype: float64
#除法：
 index1     1.5
 index2     2.0
 index3     NaN
 index33    NaN
 index4     NaN
 index44    NaN
 dtype: float64
 dataframe教程

从csv文件只读取前几行的数据

只读取前2行和指定列的数据
df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv',nrows=2,usecols=['Model','Length'])
df
#>  Model        Length
0        Integra        177
1        Legend        195

从csv文件中每隔n行来创建dataframe

每隔50行读取一行数据
df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/BostonHousing.csv', chunksize=50)
df2 = pd.DataFrame()
for chunk in df:
    # 获取series
    df2 = df2.append(chunk.iloc[0,:])
#显示前5行
print(df2.head())
#>                   crim    zn  indus chas                  nox     rm   age  \
0              0.21977   0.0   6.91    0  0.44799999999999995  5.602  62.0   
1               0.0686   0.0   2.89    0                0.445  7.416  62.5   
2   2.7339700000000002   0.0  19.58    0                0.871  5.597  94.9   
3               0.0315  95.0   1.47    0  0.40299999999999997  6.975  15.3   
4  0.19072999999999998  22.0   5.86    0                0.431  6.718  17.5   
       dis rad  tax ptratio       b  lstat  medv  
0  6.0877   3  233    17.9   396.9   16.2  19.4  
1  3.4952   2  276    18.0   396.9   6.19  33.2  
2  1.5257   5  403    14.7  351.85  21.45  15.4  
3  7.6534   3  402    17.0   396.9   4.56  34.9  
4  7.8265   7  330    19.1  393.74   6.56  26.2

改变导入csv文件的列值
改变列名‘medv’的值，当列值≤25时，赋值为‘Low’；列值>25时，赋值为‘High’.

使用converters参数，改变medv列的值
df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/BostonHousing.csv',
                 converters={'medv': lambda x: 'High' if float(x) > 25 else 'Low'})
print(df.head())
#>         b  lstat  medv
0  396.90   4.98   Low  
1  396.90   9.14   Low  
2  392.83   4.03  High  
3  394.63   2.94  High  
4  396.90   5.33  High

从csv文件导入指定的列

导入指定的列：crim和medv
df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/BostonHousing.csv', usecols=['crim', 'medv'])
打印前四行dataframe信息
print(df.head())
#>       crim  medv
0  0.00632  24.0
1  0.02731  21.6
2  0.02729  34.7
3  0.03237  33.4
4  0.06905  36.2

得到dataframe的行，列，每一列的类型和相应的描述统计信息

df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')
打印dataframe的行和列
print(df.shape)
打印dataframe每列元素的类型显示前5行
print(df.dtypes.head())
统计各类型的数目,方法1
print(df.get_dtype_counts())
统计各类型的数目,方法2
print(df.dtypes.value_counts())
描述每列的统计信息，如std，四分位数等
df_stats = df.describe()
dataframe转化数组
df_arr = df.values
数组转化为列表
df_list = df.values.tolist()
#>        (93, 27)
 Manufacturer     object
 Model            object
 Type             object
 Min.Price       float64
 Price           float64
 dtype: object
 float64    18
object      9
 dtype: int64

获取给定条件的行和列

import numpy as np
df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')
print(df)
获取最大值的行和列
row, col = np.where(df.values == np.max(df.Price))
行和列获取最大值
print(df.iat[row[0], col[0]])
df.iloc[row[0], col[0]]
行索引和列名获取最大值
df.at[row[0], 'Price']
df.get_value(row[0], 'Price')
#>        61.9

重命名dataframe的特定列

df1 = pd.DataFrame(data=np.array([[18,50],[19,51],[20,55]]),index=['man1','man2','man3'],columns=['age','weight'])
print(df1)
修改列名
print("\nchange columns :\n")
#方法1
df1.rename(columns={'weight':'stress'})
#方法2
df1.columns.values[1] = 'stress'
print(df1)
#>       age  weight
 man1   18      50
 man2   19      51
 man3   20      55
 change columns :
       age  stress
 man1   18      50
 man2   19      51
 man3   20      55

检查dataframe中是否有缺失值

df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')
若有缺失值，则为Ture
df.isnull().values.any()
#>        True

统计dataframe的每列中缺失值的个数

df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')
获取每列的缺失值个数
n_missings_each_col = df.apply(lambda x: x.isnull().sum())
print(n_missings_each_col.head())
#>        Manufacturer    4
 Model           1
 Type            3
 Min.Price       7
 Price           2
 dtype: int64

用平均值替换相应列的缺失值

df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv',nrows=10)
print(df[['Min.Price','Max.Price']].head())
平均值替换缺失值
df_out = df[['Min.Price', 'Max.Price']] = df[['Min.Price', 'Max.Price']].apply(lambda x: x.fillna(x.mean()))
print(df_out.head())
#>    Min.Price  Max.Price
0       12.9       18.8
1       29.2       38.7
2       25.9       32.3
3        NaN       44.6
4        NaN        NaN
#>    Min.Price  Max.Price
0       12.9       18.8
1       29.2       38.7
2       25.9       32.3
3       23.0       44.6
4       23.0       29.9

用全局变量作为apply函数的附加参数处理指定的列

df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')
print(df[['Min.Price', 'Max.Price']].head())
全局变量
d = {'Min.Price': np.nanmean, 'Max.Price': np.nanmedian}
列名Min.Price的缺失值用平均值代替，Max.Price的缺失值用中值代替
df[['Min.Price', 'Max.Price']] = df[['Min.Price', 'Max.Price']].apply(lambda x, d: x.fillna(d
[x.name](x)
), args=(d, ))
print(df[['Min.Price', 'Max.Price']].head())
#>    Min.Price  Max.Price
0       12.9       18.8
1       29.2       38.7
2       25.9       32.3
3        NaN       44.6
4        NaN        NaN
#>    Min.Price  Max.Price
0  12.900000      18.80
1  29.200000      38.70
2  25.900000      32.30
3  17.118605      44.60
4  17.118605      19.15

以dataframe的形式选择特定的列

df = pd.DataFrame(np.arange(20).reshape(-1, 5), columns=list('abcde'))
print(df)
以dataframe的形式选择特定的列
type(df[['a']])
type(df.loc[:, ['a']])
print(type(df.iloc[:, [0]]))
以series的形式选择特定的列
type(df.a)
type(df['a'])
type(df.loc[:, 'a'])
print(type(df.iloc[:, 1]))
#>

改变dataframe中的列顺序

df = pd.DataFrame(np.arange(20).reshape(-1, 5), columns=list('abcde'))
print(df)
交换col1和col2
def switch_columns(df, col1=None, col2=None):
    colnames = df.columns.tolist()
    i1, i2 = colnames.index(col1), colnames.index(col2)
    colnames[i2], colnames[i1] = colnames[i1], colnames[i2]
    return df[colnames]
df1 = switch_columns(df, 'a', 'c')
print(df1)
#>     a   b   c   d   e
0   0   1   2   3   4
1   5   6   7   8   9
2  10  11  12  13  14
3  15  16  17  18  19
#>     c   b   a   d   e
0   2   1   0   3   4
1   7   6   5   8   9
2  12  11  10  13  14
3  17  16  15  18  19

格式化dataframe的值

df = pd.DataFrame(np.random.random(4)**10, columns=['random'])
print(df)
显示小数点后四位
df.apply(lambda x: '%.4f' % x, axis=1)
print(df)
#>          random
0  3.539348e-04
1  3.864140e-10
2  2.973575e-02
3  1.414061e-01
#>          random
0  3.539348e-04
1  3.864140e-10
2  2.973575e-02
3  1.414061e-01

将dataframe中的所有值以百分数的格式表示

df = pd.DataFrame(np.random.random(4), columns=['random'])
格式化为小数点后两位的百分数
out = df.style.format({
    'random': '{0:.2%}'.format,
})
out
#>  random
0        48.54%
1        91.51%
2        90.83%
3        20.45%

从dataframe中每隔n行构建dataframe

df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')
每隔20行读dataframe数据
print(df.iloc[::20, :][['Manufacturer', 'Model', 'Type']])
#>    Manufacturer    Model     Type
0         Acura  Integra    Small
20     Chrysler  LeBaron  Compact
40        Honda  Prelude   Sporty
60      Mercury   Cougar  Midsize
80       Subaru   Loyale    Small

得到列中前n个最大值对应的索引

df = pd.DataFrame(np.random.randint(1, 15, 15).reshape(5,-1), columns=list('abc'))
print(df)
取'a'列前3个最大值对应的行
n = 5
df['a'].argsort()[::-1].iloc[:3]
#>     a   b   c
0   5   5   2
1  12   7   1
2   5   2  12
3   5  14  12
4   1  13  13
#>        4    1
3    3
2    2
 Name: a, dtype: int64

获得dataframe行的和大于100的最末n行索引

df = pd.DataFrame(np.random.randint(10, 40, 16).reshape(-1, 4))
print(df)
dataframe每行的和
rowsums = df.apply(np.sum, axis=1)
选取大于100的最末两行索引
last_two_rows = df.iloc[np.where(rowsums > 100)[0][-2:], :]
nline = np.where(rowsums > 100)[0][-2:]
nline
#>     0   1   2   3
0  19  34  15  12
1  38  35  14  26
2  39  32  18  20
3  28  27  36  38
#>        array([2, 3], dtype=int64)

从series中查找异常值并赋值

ser = pd.Series(np.logspace(-2, 2, 30))
小于low_per分位的数赋值为low，大于low_per分位的数赋值为high
def cap_outliers(ser, low_perc, high_perc):
    low, high = ser.quantile([low_perc, high_perc])
    print(low_perc, '%ile: ', low, '|', high_perc, '%ile: ', high)
    ser[ser < low] = low
    ser[ser > high] = high
    return(ser)
capped_ser = cap_outliers(ser, .05, .95)
#>        0.05 %ile:  0.016049294076965887 | 0.95 %ile:  63.876672220183934

交换dataframe的两行

df = pd.DataFrame(np.arange(9).reshape(3, -1))
print(df)
函数
def swap_rows(df, i1, i2):
    a, b = df.iloc[i1, :].copy(), df.iloc[i2, :].copy()
    # 通过iloc换行
    df.iloc[i1, :], df.iloc[i2, :] = b, a
    return df
2和3行互换
print(swap_rows(df, 1, 2))
#>    0  1  2
0  0  1  2
1  3  4  5
2  6  7  8
#>    0  1  2
0  0  1  2
1  6  7  8
2  3  4  5

倒转dataframe的行

df = pd.DataFrame(np.arange(9).reshape(3, -1))
print(df)
方法 1
df.iloc[::-1, :]
方法 2
print(df.loc[df.index[::-1], :])
#>    0  1  2
0  0  1  2
1  3  4  5
2  6  7  8
#>    0  1  2
2  6  7  8
1  3  4  5
0  0  1  2

对分类变量进行one-hot编码

df = pd.DataFrame(np.arange(25).reshape(5,-1), columns=list('abcde'))
print(df)
对列'a'进行onehot编码
df_onehot = pd.concat([pd.get_dummies(df['a']), df[list('bcde')]], axis=1)
print(df_onehot)
#>     a   b   c   d   e
0   0   1   2   3   4
1   5   6   7   8   9
2  10  11  12  13  14
3  15  16  17  18  19
4  20  21  22  23  24
#>    0  5  10  15  20   b   c   d   e
0  1  0   0   0   0   1   2   3   4
1  0  1   0   0   0   6   7   8   9
2  0  0   1   0   0  11  12  13  14
3  0  0   0   1   0  16  17  18  19
4  0  0   0   0   1  21  22  23  24

获取dataframe行方向上最大值个数最多的列

df = pd.DataFrame(np.random.randint(1,100, 9).reshape(3, -1))
print(df)
获取每列包含行方向上最大值的个数
count_series = df.apply(np.argmax, axis=1).value_counts()
print(count_series)
输出行方向最大值个数最多的列的索引
print('Column with highest row maxes: ', count_series.index[0])
#>     0   1   2
0  46  31  34
1  38  13   6
2   1  18  15
#>统计列的最大值的个数
         0    2
         1    1
         dtype: int64
 
#>        Column with highest row maxes:  0

得到列之间最大的相关系数

df = pd.DataFrame(np.random.randint(1,100, 16).reshape(4, -1), columns=list('pqrs'), index=list('abcd'))
df
print(df)
得到四个列的相关系数
abs_corrmat = np.abs(df.corr())
print(abs_corrmat)
得到每个列名与其他列的最大相关系数
max_corr = abs_corrmat.apply(lambda x: sorted(x)[-2])
显示每列与其他列的相关系数
print('Maximum Correlation possible for each column: ', np.round(max_corr.tolist(), 2))
#>     p   q   r   s
 a  59  99   1  34
 b  89  60  97  40
 c  43  35  14   6
 d  70  59  30  53
#>           p         q         r         s
 p  1.000000  0.200375  0.860051  0.744529
 q  0.200375  1.000000  0.236619  0.438541
 r  0.860051  0.236619  1.000000  0.341399
 s  0.744529  0.438541  0.341399  1.000000
#>        Maximum Correlation possible for each column:  [0.86 0.44 0.86 0.74]

创建包含每行最小值与最大值比例的列

df = pd.DataFrame(np.random.randint(1,100, 9).reshape(3, -1))
print(df)
方法1：axis=1表示行方向，
min_by_max = df.apply(lambda x: np.min(x)/np.max(x), axis=1)
方法2
min_by_max = np.min(df, axis=1)/np.max(df, axis=1)
min_by_max
#>     0   1   2
0  81  68  59
1  45  73  23
2  20  22  69
 
#>        0    0.728395
1    0.315068
2    0.289855
 dtype: float64

创建包含每行第二大值的列

df = pd.DataFrame(np.random.randint(1,100, 9).reshape(3, -1))
print(df)
行方向上取第二大的值组成series
out = df.apply(lambda x: x.sort_values().unique()[-2], axis=1)
构建dataframe新的列
df['penultimate'] = out
print(df)
#>     0   1   2
0  28  77   1
1  43  19  69
2  29  30  72
#>     0   1   2  penultimate
0  28  77   1           28
1  43  19  69           43
2  29  30  72           30

归一化dataframe的所有列

df = pd.DataFrame(np.random.randint(1,100, 80).reshape(8, -1))
正态分布归一化
out1 = df.apply(lambda x: ((x - x.mean())/x.std()).round(2))
print('Solution Q1\n',out1)
线性归一化
out2 = df.apply(lambda x: ((x.max() - x)/(x.max() - x.min())).round(2))
print('Solution Q2\n', out2)

计算每一行与下一行的相关性

df = pd.DataFrame(np.random.randint(1,100, 25).reshape(5, -1))
行与行之间的相关性
[df.iloc[i].corr(df.iloc[i+1]).round(2) for i in range(df.shape[0])[:-1]]

用0赋值dataframe的主对角线和副对角线

df = pd.DataFrame(np.random.randint(1,100, 25).reshape(5, -1))
print(df)
zhu
for i in range(df.shape[0]):
    df.iat[i, i] = 0
    df.iat[df.shape[0]-i-1, i] = 0
print(df)
#>     0   1   2   3   4
0  51  35  71  71  79
1  78  25  71  85  44
2  90  97  72  14   4
3  27  91  37  25  48
4   1  26  68  70  20
#>     0   1   2   3   4
0   0  35  71  71   0
1  78   0  71   0  44
2  90  97   0  14   4
3  27   0  37   0  48
4   0  26  68  70   0

得到按列分组的dataframe的平均值和标准差

df = pd.DataFrame({'col1': ['apple', 'banana', 'orange'] * 2,
                   'col2': np.random.randint(0,15,6),
                   'col3': np.random.randint(0, 15, 6)})
print(df)
按列col1分组后的平均值
df_grouped_mean = df.groupby(['col1']).mean()
print(df_grouped_mean)
按列col1分组后的标准差
df_grouped_std = df.groupby(['col1']).mean()
print(df_grouped_std)
#>      col1  col2  col3
0   apple     2    14
1  banana    11     8
2  orange     8    10
3   apple     5     2
4  banana     6    12
5  orange    11    13
#>         col2  col3
 col1

 apple    3.5   8.0
 banana   8.5  10.0
 orange   9.5  11.5
#>         col2  col3
 col1

 apple    3.5   8.0
 banana   8.5  10.0
 orange   9.5  11.5

得到按列分组后另一列的第n大的值

df = pd.DataFrame({'fruit': ['apple', 'banana', 'orange'] * 2,
                   'taste': np.random.rand(6),
                   'price': np.random.randint(0, 15, 6)})
print(df)
teste列按fruit分组
df_grpd = df['taste'].groupby(df.fruit)
teste列中banana元素的信息
x=df_grpd.get_group('banana')
排序并找第2大的值
s = x.sort_values().iloc[-2]
print(s)
#>     fruit     taste  price
0   apple  0.521990      7
1  banana  0.640444      0
2  orange  0.460509      9
3   apple  0.818963      4
4  banana  0.646138      7
5  orange  0.917056     12
#>        0.6404436436085967

计算分组dataframe的平均值，并将分组列保留为另一列

df = pd.DataFrame({'fruit': ['apple', 'banana', 'orange'] * 2,
                   'rating': np.random.rand(6),
                   'price': np.random.randint(0, 15, 6)})
按fruit分组后，price列的平均值，并将分组置为一列
out = df.groupby('fruit', as_index=False)['price'].mean()
print(out)
#>     fruit  price
0   apple    4.0
1  banana    6.5
2  orange   11.0

获取两列值元素相等的位置（并非索引）

df = pd.DataFrame({'fruit1': np.random.choice(['apple', 'orange', 'banana'], 3),
                    'fruit2': np.random.choice(['apple', 'orange', 'banana'], 3)})
print(df)
获取两列元素相等的行
np.where(df.fruit1 == df.fruit2)
#>    fruit1  fruit2
0   apple  banana
1   apple   apple
2  orange   apple
#>        (array([1], dtype=int64),)

创建指定列偏移后的新列

df = pd.DataFrame(np.random.randint(1, 100, 20).reshape(-1, 4), columns = list('abcd'))
创建往下偏移后的列
df['a_lag1'] = df['a'].shift(1)
创建往上偏移后的列
df['b_lead1'] = df['b'].shift(-1)
print(df)
#>     a   b   c   d  a_lag1  b_lead1
0  29  90  43  24     NaN     36.0
1  94  36  67  66    29.0     76.0
2  81  76  44  49    94.0     97.0
3  55  97  10  74    81.0     43.0
4  32  43  62  62    55.0      NaN

获得dataframe中单一值的频数

df = pd.DataFrame(np.random.randint(1, 10, 20).reshape(-1, 4), columns = list('abcd'))
统计元素值的个数
pd.value_counts(df.values.ravel())
#>        9    3
7    3
3    3
1    3
6    2
5    2
4    2
8    1
2    1
 dtype: int64

将文本拆分为两个单独的列

df = pd.DataFrame(["STD, City    State",
"33, Kolkata    West Bengal",
"44, Chennai    Tamil Nadu",
"40, Hyderabad    Telengana",
"80, Bangalore    Karnataka"], columns=['row'])
print(df)
expand=True表示以分割符把字符串分成两列
df_out = df.row.str.split(',|\t', expand=True)
获取新的列
new_header = df_out.iloc[0]
重新赋值
df_out = df_out[1:]
df_out.columns = new_header
print(df_out)
#>                           row
0          STD, City    State
1  33, Kolkata    West Bengal
2   44, Chennai    Tamil Nadu
3  40, Hyderabad    Telengana
4  80, Bangalore    Karnataka
#>        0 STD     City       State
1  33   Kolkata    West Bengal
2  44    Chennai    Tamil Nadu
3  40   Hyderabad    Telengana
4  80   Bangalore    Karnataka

构建多级索引的dataframe
利用元组（Tuple）构建多级索引，然后定义dataframe.

如何构建多级索引的dataframe
先通过元组方式构建多级索引
import numpy as np
outside = ['A','A','A','B','B','B']
inside =[1,2,3,1,2,3]
my_index = list(zip(outside,inside))
my_index
转化为pd格式的索引
my_index = pd.MultiIndex.from_tuples(my_index)
my_index
构建多级索引dataframe
df = pd.DataFrame(np.random.randn(6,2),index =my_index,columns=['fea1','fea2'])
df
多索引dataframe结果：
获取多索引dataframe的数据：

df.loc['A'].iloc[1]
#>        fea1   -0.794461
 fea2    0.882104
 Name: 2, dtype: float64
df.loc['A'].iloc[1]['fea1']
#>        -0.7944609970323794 0.23.4"}}},"align":"","language":"Python","wrap":false}},"doxcncLmWdikkZkDCE3Ka5AMglc":{"id":"doxcncLmWdikkZkDCE3Ka5AMglc","snapshot":{"type":"heading3","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+c"},"text":{"0":"3、pandas数据类型"}}},"align":"","folded":false}},"doxcncc98qTlXqr0d85X2s8eu8f":{"id":"doxcncc98qTlXqr0d85X2s8eu8f","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+w"},"text":{"0":"pandas包含两种数据类型：series和dataframe。"}}},"align":"","folded":false}},"doxcnHZC4BVE2Ji80SWstSfSigg":{"id":"doxcnHZC4BVE2Ji80SWstSfSigg","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+1t"},"text":{"0":"series是一种一维数据结构，每一个元素都带有一个索引，与一维数组的含义相似，其中索引可以为数字或字符串。series结构名称："}}},"align":"","folded":false}},"doxcn1U5V30YIrU61rpRrBRTbXd":{"id":"doxcn1U5V30YIrU61rpRrBRTbXd","snapshot":{"type":"image","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","align":"","image":{"token":"KCiPbzPaJoGDiJx6hQScODfBnM9","mimeType":"image/png","size":4980,"scale":1,"width":218,"height":155,"name":"Vg9dbbWIioMxS2xpz9gc9ACxnfg.png","crop":[0,0,0,0],"rotation":0,"caption":{"text":{"apool":{"nextNum":0,"numToAttrib":null},"initialAttributedTexts":{"attribs":null,"text":null}}}}}},"doxcnswLs24vvG3JP16ZiXklNre":{"id":"doxcnswLs24vvG3JP16ZiXklNre","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+1p"},"text":{"0":"dataframe是一种二维数据结构，数据以表格形式（与excel类似）存储，有对应的行和列。dataframe结构名称："}}},"align":"","folded":false}},"doxcnXAof7jWLxGZknAZa1bkCih":{"id":"doxcnXAof7jWLxGZknAZa1bkCih","snapshot":{"type":"image","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","align":"","image":{"token":"Eea7bQZ5goT8Six1wVjcW2KcnMR","mimeType":"image/webp","size":4618,"scale":1,"width":216,"height":136,"name":"XPGrbEMkvojIqZxiCUbcKUcBnSf.webp","crop":[0,0,0,0],"rotation":0,"caption":{"text":{"apool":{"nextNum":0,"numToAttrib":null},"initialAttributedTexts":{"attribs":null,"text":null}}}}}},"doxcnMju63hRn8ngF31l6Fb5Vxh":{"id":"doxcnMju63hRn8ngF31l6Fb5Vxh","snapshot":{"type":"heading3","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+b"},"text":{"0":"4、series教程 "}}},"align":"","folded":false}},"doxcnIZbax2lqCv0wTAiOxgWEfe":{"id":"doxcnIZbax2lqCv0wTAiOxgWEfe","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+h"},"text":{"0":"从列表，数组，字典构建series"}}},"align":"","folded":false,"seq":"1"}},"doxcnpAsutfNcnQXrvFvFaDUuIf":{"id":"doxcnpAsutfNcnQXrvFvFaDUuIf","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|d+96*0+f"},"text":{"0":"mylist = list('abcedfghijklmnopqrstuvwxyz') # 列表\nmyarr = np.arange(26) # 数组\nmydict = dict(zip(mylist, myarr)) # 字典\n构建方法\nser1 = pd.Series(mylist)\nser2 = pd.Series(myarr)\nser3 = pd.Series(mydict)\nprint(ser3.head()) # 打印前5个数据\n#> a 0\n b 1\n c 2\n d 4\n e 3\n dtype:int64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnTRHNLSwjSx7FysZ1tNCA4e":{"id":"doxcnTRHNLSwjSx7FysZ1tNCA4e","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+p"},"text":{"0":"使series的索引列转化为dataframe的列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnzgvPKzztjypBge56vtVkZf":{"id":"doxcnzgvPKzztjypBge56vtVkZf","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|e+7x*0+a"},"text":{"0":"mylist = list('abcedfghijklmnopqrstuvwxyz')\nmyarr = np.arange(26)\nmydict = dict(zip(mylist, myarr))\nser = pd.Series(mydict)\nseries转换为dataframe\ndf = ser.to_frame()\n索引列转换为dataframe的列\ndf.reset_index(inplace=True)\nprint(df.head())\n#> index 0\n0 a 0\n1 b 1\n2 c 2\n3 e 3\n4 d 4"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnQWDbvCUgAb7TNKatSeejee":{"id":"doxcnQWDbvCUgAb7TNKatSeejee","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+l"},"text":{"0":"结合多个series组成dataframe"}}},"align":"","folded":false,"seq":"auto"}},"doxcnTiOGA2zHaksZ5uOCgMxofe":{"id":"doxcnTiOGA2zHaksZ5uOCgMxofe","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|e+8r*0+c"},"text":{"0":"构建series1\nser1 = pd.Series(list('abcedfghijklmnopqrstuvwxyz')) \n构建series2\nser2 = pd.Series(np.arange(26))\n方法1，axis=1表示列拼接，0表示行拼接\ndf = pd.concat([ser1, ser2], axis=1)\n与方法1相比，方法2设置了列名\ndf = pd.DataFrame({'col1': ser1, 'col2': ser2})\nprint(df.head())\n#> col1 col2\n0 a 0\n1 b 1\n2 c 2\n3 e 3\n4 d 4"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcntVGXD5ZeGYw179Mca8TPQb":{"id":"doxcntVGXD5ZeGYw179Mca8TPQb","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+8"},"text":{"0":"命名列索引的名称"}}},"align":"","folded":false,"seq":"auto"}},"doxcn8frvSrzlSlTVh9ac3uG9Bb":{"id":"doxcn8frvSrzlSlTVh9ac3uG9Bb","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|a+43*0+v"},"text":{"0":"ser = pd.Series(list('abcedfghijklmnopqrstuvwxyz'))\n命名索引列名称\nser.name = 'alphabets'\n显示前5行数据\nser.head()\n#> 0 a\n1 b\n2 c\n3 e\n4 d\n Name: alphabets, dtype: object"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnufiEauX5gsP79cGhaUYA6c":{"id":"doxcnufiEauX5gsP79cGhaUYA6c","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+r"},"text":{"0":"获得series对象A中不包含series对象B的元素"}}},"align":"","folded":false,"seq":"auto"}},"doxcnjScfKt3D5AzKCQxvwKWDUb":{"id":"doxcnjScfKt3D5AzKCQxvwKWDUb","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|9+4s*0+d"},"text":{"0":"ser1 = pd.Series([1, 2, 3, 4, 5])\nser2 = pd.Series([4, 5, 6, 7, 8])\n返回ser1不包含ser2的布尔型series\nser3=~ser1.isin(ser2)\n获取ser不包含ser2的元素\nser1[ser3]\n#> 0 1\n1 2\n2 3\n dtype: int64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnoRhxOpX2v7SMai3zl8Ivvd":{"id":"doxcnoRhxOpX2v7SMai3zl8Ivvd","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+m"},"text":{"0":"获得seriesA和seriesB不相同的项"}}},"align":"","folded":false,"seq":"auto"}},"doxcnCMV0TGhZ6G4RyyEj73j44f":{"id":"doxcnCMV0TGhZ6G4RyyEj73j44f","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|e+85*0+d"},"text":{"0":"ser1 = pd.Series([1, 2, 3, 4, 5])\nser2 = pd.Series([4, 5, 6, 7, 8])\n求ser1和ser2的并集\nser_u = pd.Series(np.union1d(ser1, ser2))\n求ser1和ser2的交集\nser_i = pd.Series(np.intersect1d(ser1, ser2))\nser_i在ser_u的补集就是ser1和ser2不相同的项\nser_u[~ser_u.isin(ser_i)]\n#> 0 1\n1 2\n2 3\n5 6\n6 7\n7 8\n dtype: int64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnM6WCByb3uUqio3AYsdPWyf":{"id":"doxcnM6WCByb3uUqio3AYsdPWyf","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+f"},"text":{"0":"获得数值series的四分位值"}}},"align":"","folded":false,"seq":"auto"}},"doxcnUANTw2hmXue6QqizEizWKh":{"id":"doxcnUANTw2hmXue6QqizEizWKh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|6+4p*0+23"},"text":{"0":"设置随机数种子\nstate = np.random.RandomState(100)\n从均值为5标准差为25的正态分布随机抽取5个点构成series\nser = pd.Series(state.normal(10, 5, 25))\n求ser的四分位数\nnp.percentile(ser, q=[0, 25, 50, 75, 100])\n#> array([ 1.25117263, 7.70986507, 10.92259345, 13.36360403, 18.0949083 ])"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnsNOyxbu7zkVsNlATUZt0ub":{"id":"doxcnsNOyxbu7zkVsNlATUZt0ub","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+h"},"text":{"0":"获得series中单一项的频率计数"}}},"align":"","folded":false,"seq":"auto"}},"doxcnIdLKJwKuNEv698aFfXNlIt":{"id":"doxcnIdLKJwKuNEv698aFfXNlIt","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|b+5g*0+d"},"text":{"0":"#从0~7随机抽取30个列表值，组成series\nser = pd.Series(np.take(list('abcdefgh'), np.random.randint(8, size=30)))\n对该series进行计数\nser.value_counts()\n#> d 8\n g 6\n b 6\n a 5\n e 2\n h 2\n f 1\n dtype: int64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn2K7O1vuEhkG1DO4B6Ddm6e":{"id":"doxcn2K7O1vuEhkG1DO4B6Ddm6e","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+w"},"text":{"0":"保留series中前两个频次最多的项，其他项替换为‘other’"}}},"align":"","folded":false,"seq":"auto"}},"doxcn6ZTmzwk6veIzfDktgd0Ssh":{"id":"doxcn6ZTmzwk6veIzfDktgd0Ssh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|i+9d*0+e"},"text":{"0":"np.random.RandomState(100)\n从1~4均匀采样12个点组成series\nser = pd.Series(np.random.randint(1, 5, [12]))\n除前两行索引对应的值不变，后几行索引对应的值为Other\nser[~ser.isin(ser.value_counts().index[:2])] = 'Other'\nser\n#> 0 Other\n1 4\n2 2\n3 2\n4 4\n5 Other\n6 Other\n7 Other\n8 4\n9 4\n10 4\n11 2\n dtype: object"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnSc7yZ2BuuIYzM8pT5KZviw":{"id":"doxcnSc7yZ2BuuIYzM8pT5KZviw","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+k"},"text":{"0":"对数值series分成10个相同数目的组"}}},"align":"","folded":false,"seq":"auto"}},"doxcnzNadbox0muCzQStBlkWFBd":{"id":"doxcnzNadbox0muCzQStBlkWFBd","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+13"},"text":{"0":"换个角度理解，对数值series离散化成10个类别（categorical）值"}}},"align":"","folded":false}},"doxcnyZOFv3gGqZBlNVMe9UqOih":{"id":"doxcnyZOFv3gGqZBlNVMe9UqOih","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|b+7q*0+23"},"text":{"0":"ser = pd.Series(np.random.random(20))\n离散化10个类别值,只显示前5行的数据\npd.qcut(ser, q=[0, .10, .20, .3, .4, .5, .6, .7, .8, .9, 1],\n labels=['1st', '2nd', '3rd', '4th', '5th', '6th', '7th', '8th', '9th', '10th']).head()\n#>\n0 3rd\n1 1st\n2 6th\n3 6th\n4 9th\ndtype: category\nCategories (10, object): [1st < 2nd < 3rd < 4th ... 7th < 8th < 9th < 10th]"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnaI5I9koRMU2sa6FOXP67Ad":{"id":"doxcnaI5I9koRMU2sa6FOXP67Ad","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+p"},"text":{"0":"使numpy数组转化为给定形状的dataframe"}}},"align":"","folded":false,"seq":"auto"}},"doxcn5621kGVGfpskfAUSrh1Vjd":{"id":"doxcn5621kGVGfpskfAUSrh1Vjd","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|b+6s*0+g"},"text":{"0":"ser = pd.Series(np.random.randint(1, 10, 35))\nserier类型转换numpy类型，然后重构\ndf = pd.DataFrame(ser.values.reshape(7,5))\nprint(df)\n#> 0 1 2 3 4\n0 1 2 1 2 5\n1 1 2 4 5 2\n2 1 3 3 2 8\n3 8 6 4 9 6\n4 2 1 1 8 5\n5 3 2 8 5 6\n6 1 5 5 4 6"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnhu5fOuL30Y4TPy7GTLj6Bg":{"id":"doxcnhu5fOuL30Y4TPy7GTLj6Bg","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+i"},"text":{"0":"找到series的值是3的倍数的位置"}}},"align":"","folded":false,"seq":"auto"}},"doxcnDi7S0HztqExvkWEsgmZVqh":{"id":"doxcnDi7S0HztqExvkWEsgmZVqh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|f+5l*0+d"},"text":{"0":"ser = pd.Series(np.random.randint(1, 10, 7))\nprint(ser)\n获取值是3倍数的索引\nnp.argwhere(ser % 3==0)\n#> 0 6\n1 8\n2 6\n3 7\n4 6\n5 2\n6 4\n dtype: int64\n \n#> array([[0],\n [2],\n [4]])"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnjZK8AZIRC0m3fcEKIaHWTb":{"id":"doxcnjZK8AZIRC0m3fcEKIaHWTb","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+n"},"text":{"0":"获取series中给定索引的元素（items）"}}},"align":"","folded":false,"seq":"auto"}},"doxcnIbwFpZB2Zucx30Pzbfo27c":{"id":"doxcnIbwFpZB2Zucx30Pzbfo27c","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|9+4a*0+e"},"text":{"0":"ser = pd.Series(list('abcdefghijklmnopqrstuvwxyz'))\nindex = [0, 4, 8, 14, 20]\n获取指定索引的元素\nser.take(index)\n#> 0 a\n4 e\n8 i\n14 o\n20 u\n dtype: object"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn4Hl615SXgJOfgej9h3QQ0f":{"id":"doxcn4Hl615SXgJOfgej9h3QQ0f","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+e"},"text":{"0":"垂直和水平的拼接series"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnK5gaYyuBZrfEYZlOFTKeJh":{"id":"doxcnK5gaYyuBZrfEYZlOFTKeJh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|c+5g*0+7"},"text":{"0":"ser1 = pd.Series(range(5))\nser2 = pd.Series(list('abcde'))\n垂直拼接\ndf = pd.concat([ser1, ser2], axis=0)\n水平拼接\ndf = pd.concat([ser1, ser2], axis=1)\nprint(df)\n#> 0 1\n0 0 a\n1 1 b\n2 2 c\n3 3 d\n4 4 e"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnJXasa1u6tB8EuNdHHJ6emg":{"id":"doxcnJXasa1u6tB8EuNdHHJ6emg","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+s"},"text":{"0":"获取series对象A中包含series对象B元素的位置"}}},"align":"","folded":false,"seq":"auto"}},"doxcnnykCiu3OrIZOLvUqLh3wKg":{"id":"doxcnnykCiu3OrIZOLvUqLh3wKg","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|7+5o*0+f"},"text":{"0":"ser1必须包含ser2，否则会报错\nser1 = pd.Series([10, 9, 6, 5, 3, 1, 12, 8, 13])\nser2 = pd.Series([1, 3, 10, 13])\n方法 1\n[np.where(i == ser1)[0].tolist()[0] for i in ser2]\n方法 2\n[pd.Index(ser1).get_loc(i) for i in ser2]\n#> [5, 4, 0, 8]"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn6CEeMJyPy9K9NnA5ryfTWg":{"id":"doxcn6CEeMJyPy9K9NnA5ryfTWg","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+e"},"text":{"0":"计算series之间的均方差"}}},"align":"","folded":false,"seq":"auto"}},"doxcn0ML0hiLMnajFcK714dhnid":{"id":"doxcn0ML0hiLMnajFcK714dhnid","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|4+31*0+m"},"text":{"0":"truth = pd.Series(range(10))\npred = pd.Series(range(10)) + np.random.random(10)\n均方差\nnp.mean((truth-pred)**2)\n#> 0.25508722434194103"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcndp7mEwLbDg7dYegIkjWdnd":{"id":"doxcndp7mEwLbDg7dYegIkjWdnd","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+j"},"text":{"0":"使series中每个元素的首字母为大写"}}},"align":"","folded":false,"seq":"auto"}},"doxcnZnzRDZS1hwIE5GJ8MMp8eg":{"id":"doxcnZnzRDZS1hwIE5GJ8MMp8eg","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|c+6o*0+e"},"text":{"0":"series的元素为str类型\nser = pd.Series(['how', 'to', 'kick', 'ass?'])\n方法 1\nser.map(lambda x: x.title())\n方法 2 ，字符串相加\nser.map(lambda x: x[0].upper() + x[1:])\n方法 3\npd.Series([i.title() for i in ser])\n#> 0 How\n1 To\n2 Kick\n3 Ass?\n dtype: object"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnwJPX062FrA8Gw15LXfcyrc":{"id":"doxcnwJPX062FrA8Gw15LXfcyrc","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+j"},"text":{"0":"计算series中每个元素的字符串长度"}}},"align":"","folded":false,"seq":"auto"}},"doxcn4dC69t4oNWCx3dPI5FQlqf":{"id":"doxcn4dC69t4oNWCx3dPI5FQlqf","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|7+36*0+d"},"text":{"0":"ser = pd.Series(['how', 'to', 'kick', 'ass?'])\n方法\nser.map(lambda x: len(x))\n#> 0 3\n1 2\n2 4\n3 4\n dtype: int64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcngSrF1MHJESrXvsLXcmyAVa":{"id":"doxcngSrF1MHJESrXvsLXcmyAVa","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"计算series的一阶导和二阶导"}}},"align":"","folded":false,"seq":"auto"}},"doxcn56NDlN9nww0ZawWiSxfkpf":{"id":"doxcn56NDlN9nww0ZawWiSxfkpf","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|6+55*0+15"},"text":{"0":"ser = pd.Series([1, 3, 6, 10, 15, 21, 27, 35])\n求一阶导并转化为列表类型\nprint(ser.diff().tolist())\n求二阶导并转化为列表类型\nprint(ser.diff().diff().tolist())\n#> [nan, 2.0, 3.0, 4.0, 5.0, 6.0, 6.0, 8.0]\n [nan, nan, 1.0, 1.0, 1.0, 1.0, 0.0, 2.0]"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnmnvJtVb5xifEJZMP28ycOd":{"id":"doxcnmnvJtVb5xifEJZMP28ycOd","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+m"},"text":{"0":"将一系列日期字符串转换为timeseries"}}},"align":"","folded":false,"seq":"auto"}},"doxcnbb2yY0wCcdRTYOfZVp4zlh":{"id":"doxcnbb2yY0wCcdRTYOfZVp4zlh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|8+7t*0+m"},"text":{"0":"ser = pd.Series(['01 Jan 2010', '02-02-2011', '20120303', '2013/04/04', '2014-05-05', '2015-06-06T12:20'])\npd.to_datetime(ser)\n#> 0 2010-01-01 00:00:00\n1 2011-02-02 00:00:00\n2 2012-03-03 00:00:00\n3 2013-04-04 00:00:00\n4 2014-05-05 00:00:00\n5 2015-06-06 12:20:00\n dtype: datetime64[ns]"}}},"align":"","language":"Python","wrap":false}},"doxcnpIgwqUowns3ly5JhhybMDd":{"id":"doxcnpIgwqUowns3ly5JhhybMDd","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+n"},"text":{"0":"从一个series中获取至少包含两个元音的元素"}}},"align":"","folded":false,"seq":"auto"}},"doxcnkBpkcWb4h5aeSoRmeZZ2ub":{"id":"doxcnkBpkcWb4h5aeSoRmeZZ2ub","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|9+83*0+e"},"text":{"0":"ser = pd.Series(['Apple', 'Orange', 'Plan', 'Python', 'Money'])\n方法\nfrom collections import Counter\nCounter是一个类字典类型，键是元素值，值是元素出现的次数,满足条件的元素返回True\nmask = ser.map(lambda x: sum([Counter(x.lower()).get(i, 0) for i in list('aeiou')]) >= 2)\nser[mask]\n#> 0 Apple\n1 Orange\n4 Money\n dtype: object"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnn5kOt9nLdqzpKJhRO1NEsk":{"id":"doxcnn5kOt9nLdqzpKJhRO1NEsk","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+p"},"text":{"0":"计算根据另一个series分组后的series均值"}}},"align":"","folded":false,"seq":"auto"}},"doxcnXYaC5zCdivW7gtNo0IG8yb":{"id":"doxcnXYaC5zCdivW7gtNo0IG8yb","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|c+c6*0+f"},"text":{"0":"fruit = pd.Series(np.random.choice(['apple', 'banana', 'carrot'], 10))\nweights = pd.Series(np.linspace(1, 10, 10))\n根据fruit对weight分组\nweightsGrouped = weights.groupby(fruit)\nprint(weightsGrouped.indices)\n对分组后series求每个索引的平均值\nweightsGrouped.mean()\n#> {'apple': array([0, 3], dtype=int64), 'banana': array([1, 2, 4, 8],\n dtype=int64), 'carrot': array([5, 6, 7, 9], dtype=int64)}\n#> apple 2.50\n banana 4.75\n carrot 7.75\n dtype: float64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnMN7OrEgBd5p3VY542KBUBe":{"id":"doxcnMN7OrEgBd5p3VY542KBUBe","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+h"},"text":{"0":"计算两个series之间的欧氏距离"}}},"align":"","folded":false,"seq":"auto"}},"doxcnyG10bN7OcZetGXaMC2Ssge":{"id":"doxcnyG10bN7OcZetGXaMC2Ssge","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|8+40*0+r"},"text":{"0":"p = pd.Series([1, 2, 3, 4, 5, 6, 7, 8, 9, 10])\nq = pd.Series([10, 9, 8, 7, 6, 5, 4, 3, 2, 1])\n方法1\nsum((p - q)\n**2)**\n.5\n方法2\nnp.linalg.norm(p-q)\n#> 18.16590212458495"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnE3JY4UP13lsWX2YAV1UEKb":{"id":"doxcnE3JY4UP13lsWX2YAV1UEKb","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"在数值series中找局部最大值"}}},"align":"","folded":false,"seq":"auto"}},"doxcnxNO5DBjsKA3hHXr4ZdsjDb":{"id":"doxcnxNO5DBjsKA3hHXr4ZdsjDb","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+f"},"text":{"0":"局部最大值对应二阶导局部最小值"}}},"align":"","folded":false}},"doxcnvzAvBDNNccupAPpapRoLde":{"id":"doxcnvzAvBDNNccupAPpapRoLde","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|6+4g*0+13"},"text":{"0":"ser = pd.Series([2, 10, 3, 4, 9, 10, 2, 7, 3])\n二阶导\ndd = np.diff(np.sign(np.diff(ser)))\n二阶导的最小值对应的值为最大值，返回最大值的索引\npeak_locs = np.where(dd == -2)[0] + 1\npeak_locs\n#> array([1, 5, 7], dtype=int64)"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnpfBtvLEbJb5Ow30wwahi6g":{"id":"doxcnpfBtvLEbJb5Ow30wwahi6g","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+d"},"text":{"0":"用最少出现的字符替换空格符"}}},"align":"","folded":false,"seq":"auto"}},"doxcnW96M1lVwQevxGHIdAwsBob":{"id":"doxcnW96M1lVwQevxGHIdAwsBob","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|i+7z*0+t"},"text":{"0":"my_str = 'dbc deb abed gade'\n方法\nser = pd.Series(list('dbc deb abed gade'))\n统计元素的频数\nfreq = ser.value_counts()\nprint(freq)\n求最小频数的字符\nleast_freq = freq.dropna().index[-1]\n替换\n\"\".join(ser.replace(' ', least_freq))\n#> d 4\n 3\n b 3\n e 3\n a 2\n c 1\n g 1\n dtype: int64\n#> 'dbcgdebgabedggade'"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnFe2PlkgVgS0CpMpgLUhKif":{"id":"doxcnFe2PlkgVgS0CpMpgLUhKif","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"计算数值series的自相关系数"}}},"align":"","folded":false,"seq":"auto"}},"doxcnmRLDwpeCuVR4c3XXuP9pMf":{"id":"doxcnmRLDwpeCuVR4c3XXuP9pMf","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|7+9h*0+18"},"text":{"0":"ser = pd.Series(np.arange(20) + np.random.normal(1, 10, 20))\n求series的自相关系数，i为偏移量\nautocorrelations = [ser.autocorr(i).round(2) for i in range(11)]\nprint(autocorrelations[1:])\n选择最大的偏移量\nprint('Lag having highest correlation: ', np.argmax(np.abs(autocorrelations[1:]))+1)\n#> [0.33, 0.41, 0.48, 0.01, 0.21, 0.16, -0.11, 0.05, 0.34, -0.24]\n#> Lag having highest correlation: 3"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnShQpxjmFWcKKj0ZbV8ymQc":{"id":"doxcnShQpxjmFWcKKj0ZbV8ymQc","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+f"},"text":{"0":"对series进行算术运算操作"}}},"align":"","folded":false,"seq":"auto"}},"doxcnySbPoaoBTkdhopxWtexQBd":{"id":"doxcnySbPoaoBTkdhopxWtexQBd","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|x+it*0+c"},"text":{"0":"如何对series之间进行算法运算\nimport pandas as pd\nseries1 = pd.Series([3,4,4,4],['index1','index2','index3','index4'])\nseries2 = pd.Series([2,2,2,2],['index1','index2','index33','index44'])\n加法\nseries_add = series1 + series2\nprint(series_add)\n减法\nseries_minus = series1 - series2\nseries_minus\n乘法\nseries_multi = series1 * series2\nseries_multi\n除法\nseries_div = series1/series2\nseries_div\nseries是基于索引进行算数运算操作的，pandas会根据索引对数据进行运算，若series之间有不同的索引，对应的值就为Nan。结果如下：\n#加法：\n index1 5.0\n index2 6.0\n index3 NaN\n index33 NaN\n index4 NaN\n index44 NaN\n dtype: float64\n#除法：\n index1 1.5\n index2 2.0\n index3 NaN\n index33 NaN\n index4 NaN\n index44 NaN\n dtype: float64\n dataframe教程"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnjhca9AVbknM7gRhzPc43we":{"id":"doxcnjhca9AVbknM7gRhzPc43we","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":0,"numToAttrib":{}},"initialAttributedTexts":{"attribs":{"0":""},"text":{"0":""}}},"align":"","folded":false}},"doxcnBbXCVQeOGUtZmhRQzdBjji":{"id":"doxcnBbXCVQeOGUtZmhRQzdBjji","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+f"},"text":{"0":"从csv文件只读取前几行的数据"}}},"align":"","folded":false,"seq":"auto"}},"doxcnY6TrasVJsYBNNMeq4KlORS":{"id":"doxcnY6TrasVJsYBNNMeq4KlORS","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|5+5i*0+q"},"text":{"0":"只读取前2行和指定列的数据\ndf = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv',nrows=2,usecols=['Model','Length'])\ndf\n#> Model Length\n0 Integra 177\n1 Legend 195"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnZ5Zpn4NoVc3lFU8KMQOlpE":{"id":"doxcnZ5Zpn4NoVc3lFU8KMQOlpE","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+n"},"text":{"0":"从csv文件中每隔n行来创建dataframe"}}},"align":"","folded":false,"seq":"auto"}},"doxcnNLqv4k0rbPpING3fmLw2gg":{"id":"doxcnNLqv4k0rbPpING3fmLw2gg","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|j+qo*0+1b"},"text":{"0":"每隔50行读取一行数据\ndf = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/BostonHousing.csv', chunksize=50)\ndf2 = pd.DataFrame()\nfor chunk in df:\n # 获取series\n df2 = df2.append(chunk.iloc[0,:])\n#显示前5行\nprint(df2.head())\n#> crim zn indus chas nox rm age \\\n0 0.21977 0.0 6.91 0 0.44799999999999995 5.602 62.0 \n1 0.0686 0.0 2.89 0 0.445 7.416 62.5 \n2 2.7339700000000002 0.0 19.58 0 0.871 5.597 94.9 \n3 0.0315 95.0 1.47 0 0.40299999999999997 6.975 15.3 \n4 0.19072999999999998 22.0 5.86 0 0.431 6.718 17.5 \n dis rad tax ptratio b lstat medv \n0 6.0877 3 233 17.9 396.9 16.2 19.4 \n1 3.4952 2 276 18.0 396.9 6.19 33.2 \n2 1.5257 5 403 14.7 351.85 21.45 15.4 \n3 7.6534 3 402 17.0 396.9 4.56 34.9 \n4 7.8265 7 330 19.1 393.74 6.56 26.2"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnVSeSN29rmGxOHh91gXtG2b":{"id":"doxcnVSeSN29rmGxOHh91gXtG2b","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+c"},"text":{"0":"改变导入csv文件的列值"}}},"align":"","folded":false,"seq":"auto"}},"doxcnBOt3E2IPjglm9bgS3ARGte":{"id":"doxcnBOt3E2IPjglm9bgS3ARGte","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+1b"},"text":{"0":"改变列名‘medv’的值，当列值≤25时，赋值为‘Low’；列值>25时，赋值为‘High’."}}},"align":"","folded":false}},"doxcnHVnPDKiU7yv8BTCg9h5qPe":{"id":"doxcnHVnPDKiU7yv8BTCg9h5qPe","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|9+9o*0+m"},"text":{"0":"使用converters参数，改变medv列的值\ndf = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/BostonHousing.csv',\n converters={'medv': lambda x: 'High' if float(x) > 25 else 'Low'})\nprint(df.head())\n#> b lstat medv\n0 396.90 4.98 Low \n1 396.90 9.14 Low \n2 392.83 4.03 High \n3 394.63 2.94 High \n4 396.90 5.33 High"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnq88mUIGE4KxyihW5RDPI61":{"id":"doxcnq88mUIGE4KxyihW5RDPI61","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+c"},"text":{"0":"从csv文件导入指定的列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnq7NY1oPLs8GZr7X639vF9e":{"id":"doxcnq7NY1oPLs8GZr7X639vF9e","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|9+79*0+g"},"text":{"0":"导入指定的列：crim和medv\ndf = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/BostonHousing.csv', usecols=['crim', 'medv'])\n打印前四行dataframe信息\nprint(df.head())\n#> crim medv\n0 0.00632 24.0\n1 0.02731 21.6\n2 0.02729 34.7\n3 0.03237 33.4\n4 0.06905 36.2"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnos8LJdTZm9JDyQcoS0wyB0":{"id":"doxcnos8LJdTZm9JDyQcoS0wyB0","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+w"},"text":{"0":"得到dataframe的行，列，每一列的类型和相应的描述统计信息"}}},"align":"","folded":false,"seq":"auto"}},"doxcnhQ8ai47rk7YI7XxTlRmcGO":{"id":"doxcnhQ8ai47rk7YI7XxTlRmcGO","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|o+fp*0+d"},"text":{"0":"df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')\n打印dataframe的行和列\nprint(df.shape)\n打印dataframe每列元素的类型显示前5行\nprint(df.dtypes.head())\n统计各类型的数目,方法1\nprint(df.get_dtype_counts())\n统计各类型的数目,方法2\nprint(df.dtypes.value_counts())\n描述每列的统计信息，如std，四分位数等\ndf_stats = df.describe()\ndataframe转化数组\ndf_arr = df.values\n数组转化为列表\ndf_list = df.values.tolist()\n#> (93, 27)\n Manufacturer object\n Model object\n Type object\n Min.Price float64\n Price float64\n dtype: object\n float64 18\nobject 9\n dtype: int64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnGr6WT4IBoEy31MrS2YEWKh":{"id":"doxcnGr6WT4IBoEy31MrS2YEWKh","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+a"},"text":{"0":"获取给定条件的行和列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnOi7Q28BU1nsulYmcCHOyrd":{"id":"doxcnOi7Q28BU1nsulYmcCHOyrd","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|b+8o*0+e"},"text":{"0":"import numpy as np\ndf = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')\nprint(df)\n获取最大值的行和列\nrow, col = np.where(df.values == np.max(df.Price))\n行和列获取最大值\nprint(df.iat[row[0], col[0]])\ndf.iloc[row[0], col[0]]\n行索引和列名获取最大值\ndf.at[row[0], 'Price']\ndf.get_value(row[0], 'Price')\n#> 61.9"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnADu5qHAmF21NboPXCIqDUf":{"id":"doxcnADu5qHAmF21NboPXCIqDUf","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"重命名dataframe的特定列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnsreRpfdnYnVxKfd8XxCvKh":{"id":"doxcnsreRpfdnYnVxKfd8XxCvKh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|h+bc*0+i"},"text":{"0":"df1 = pd.DataFrame(data=np.array([[18,50],[19,51],[20,55]]),index=['man1','man2','man3'],columns=['age','weight'])\nprint(df1)\n修改列名\nprint(\"\\nchange columns :\\n\")\n#方法1\ndf1.rename(columns={'weight':'stress'})\n#方法2\ndf1.columns.values[1] = 'stress'\nprint(df1)\n#> age weight\n man1 18 50\n man2 19 51\n man3 20 55\n change columns :\n age stress\n man1 18 50\n man2 19 51\n man3 20 55"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnxHedW5jNMRsSfC02ZAThng":{"id":"doxcnxHedW5jNMRsSfC02ZAThng","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+i"},"text":{"0":"检查dataframe中是否有缺失值"}}},"align":"","folded":false,"seq":"auto"}},"doxcnffYje0A03g0sRrQq8HQtYd":{"id":"doxcnffYje0A03g0sRrQq8HQtYd","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|3+3o*0+e"},"text":{"0":"df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')\n若有缺失值，则为Ture\ndf.isnull().values.any()\n#> True"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnfT18gUJC1e5lRkiS2xIgDc":{"id":"doxcnfT18gUJC1e5lRkiS2xIgDc","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+l"},"text":{"0":"统计dataframe的每列中缺失值的个数"}}},"align":"","folded":false,"seq":"auto"}},"doxcniNaO0YF7eHot9uvcOFRmIs":{"id":"doxcniNaO0YF7eHot9uvcOFRmIs","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|9+8e*0+d"},"text":{"0":"df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')\n获取每列的缺失值个数\nn_missings_each_col = df.apply(lambda x: x.isnull().sum())\nprint(n_missings_each_col.head())\n#> Manufacturer 4\n Model 1\n Type 3\n Min.Price 7\n Price 2\n dtype: int64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnytYH8AN71cB1KJXR2w6zMg":{"id":"doxcnytYH8AN71cB1KJXR2w6zMg","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+d"},"text":{"0":"用平均值替换相应列的缺失值"}}},"align":"","folded":false,"seq":"auto"}},"doxcnDkS4YLb4ejiNvIxDSwJ4e9":{"id":"doxcnDkS4YLb4ejiNvIxDSwJ4e9","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|g+fg*0+n"},"text":{"0":"df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv',nrows=10)\nprint(df[['Min.Price','Max.Price']].head())\n平均值替换缺失值\ndf_out = df[['Min.Price', 'Max.Price']] = df[['Min.Price', 'Max.Price']].apply(lambda x: x.fillna(x.mean()))\nprint(df_out.head())\n#> Min.Price Max.Price\n0 12.9 18.8\n1 29.2 38.7\n2 25.9 32.3\n3 NaN 44.6\n4 NaN NaN\n#> Min.Price Max.Price\n0 12.9 18.8\n1 29.2 38.7\n2 25.9 32.3\n3 23.0 44.6\n4 23.0 29.9"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn7ETXSUxT9qpIf49tAIGkeg":{"id":"doxcn7ETXSUxT9qpIf49tAIGkeg","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+p"},"text":{"0":"用全局变量作为apply函数的附加参数处理指定的列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnJ9wLnv4yuW3WxXdeaCHeUh":{"id":"doxcnJ9wLnv4yuW3WxXdeaCHeUh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|k+iu*0+n"},"text":{"0":"df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')\nprint(df[['Min.Price', 'Max.Price']].head())\n全局变量\nd = {'Min.Price': np.nanmean, 'Max.Price': np.nanmedian}\n列名Min.Price的缺失值用平均值代替，Max.Price的缺失值用中值代替\ndf[['Min.Price', 'Max.Price']] = df[['Min.Price', 'Max.Price']].apply(lambda x, d: x.fillna(d\n[x.name](x)\n), args=(d, ))\nprint(df[['Min.Price', 'Max.Price']].head())\n#> Min.Price Max.Price\n0 12.9 18.8\n1 29.2 38.7\n2 25.9 32.3\n3 NaN 44.6\n4 NaN NaN\n#> Min.Price Max.Price\n0 12.900000 18.80\n1 29.200000 38.70\n2 25.900000 32.30\n3 17.118605 44.60\n4 17.118605 19.15"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn7b3mdmtAABCkuyuGXkYfId":{"id":"doxcn7b3mdmtAABCkuyuGXkYfId","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+j"},"text":{"0":"以dataframe的形式选择特定的列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnnTBI5dltwM73MHHNIl4Ibh":{"id":"doxcnnTBI5dltwM73MHHNIl4Ibh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|c+8j*0+z"},"text":{"0":"df = pd.DataFrame(np.arange(20).reshape(-1, 5), columns=list('abcde'))\nprint(df)\n以dataframe的形式选择特定的列\ntype(df[['a']])\ntype(df.loc[:, ['a']])\nprint(type(df.iloc[:, [0]]))\n以series的形式选择特定的列\ntype(df.a)\ntype(df['a'])\ntype(df.loc[:, 'a'])\nprint(type(df.iloc[:, 1]))\n#> \n"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnY31bnFKqQazJFWmPQjJtTd":{"id":"doxcnY31bnFKqQazJFWmPQjJtTd","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"改变dataframe中的列顺序"}}},"align":"","folded":false,"seq":"auto"}},"doxcnzPg7cUvsARaTQDnAGTTloc":{"id":"doxcnzPg7cUvsARaTQDnAGTTloc","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|j+fo*0+l"},"text":{"0":"df = pd.DataFrame(np.arange(20).reshape(-1, 5), columns=list('abcde'))\nprint(df)\n交换col1和col2\ndef switch_columns(df, col1=None, col2=None):\n colnames = df.columns.tolist()\n i1, i2 = colnames.index(col1), colnames.index(col2)\n colnames[i2], colnames[i1] = colnames[i1], colnames[i2]\n return df[colnames]\ndf1 = switch_columns(df, 'a', 'c')\nprint(df1)\n#> a b c d e\n0 0 1 2 3 4\n1 5 6 7 8 9\n2 10 11 12 13 14\n3 15 16 17 18 19\n#> c b a d e\n0 2 1 0 3 4\n1 7 6 5 8 9\n2 12 11 10 13 14\n3 17 16 15 18 19"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnXBBBAM1JPSTzPt22lPBkZc":{"id":"doxcnXBBBAM1JPSTzPt22lPBkZc","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+e"},"text":{"0":"格式化dataframe的值"}}},"align":"","folded":false,"seq":"auto"}},"doxcnzmi8YLHvbV1Sm8HFaxudJf":{"id":"doxcnzmi8YLHvbV1Sm8HFaxudJf","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|e+7t*0+f"},"text":{"0":"df = pd.DataFrame(np.random.random(4)**10, columns=['random'])\nprint(df)\n显示小数点后四位\ndf.apply(lambda x: '%.4f' % x, axis=1)\nprint(df)\n#> random\n0 3.539348e-04\n1 3.864140e-10\n2 2.973575e-02\n3 1.414061e-01\n#> random\n0 3.539348e-04\n1 3.864140e-10\n2 2.973575e-02\n3 1.414061e-01"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnEe0kvQVwnd7bHsQ1DkQLUe":{"id":"doxcnEe0kvQVwnd7bHsQ1DkQLUe","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+o"},"text":{"0":"将dataframe中的所有值以百分数的格式表示"}}},"align":"","folded":false,"seq":"auto"}},"doxcn5A91Y5O3Nt3kUBML0RQVwC":{"id":"doxcn5A91Y5O3Nt3kUBML0RQVwC","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|a+5g*0+f"},"text":{"0":"df = pd.DataFrame(np.random.random(4), columns=['random'])\n格式化为小数点后两位的百分数\nout = df.style.format({\n 'random': '{0:.2%}'.format,\n})\nout\n#> random\n0 48.54%\n1 91.51%\n2 90.83%\n3 20.45%"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnlVWqEr1twoRfj4wVLeURXb":{"id":"doxcnlVWqEr1twoRfj4wVLeURXb","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+q"},"text":{"0":"从dataframe中每隔n行构建dataframe"}}},"align":"","folded":false,"seq":"auto"}},"doxcnTDMsRcNOikYW9m1S6nfsDu":{"id":"doxcnTDMsRcNOikYW9m1S6nfsDu","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|8+9k*0+x"},"text":{"0":"df = pd.read_csv('https://raw.githubusercontent.com/selva86/datasets/master/Cars93_miss.csv')\n每隔20行读dataframe数据\nprint(df.iloc[::20, :][['Manufacturer', 'Model', 'Type']])\n#> Manufacturer Model Type\n0 Acura Integra Small\n20 Chrysler LeBaron Compact\n40 Honda Prelude Sporty\n60 Mercury Cougar Midsize\n80 Subaru Loyale Small"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcncPj13pCOHRnrFuiFRQr7cc":{"id":"doxcncPj13pCOHRnrFuiFRQr7cc","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+f"},"text":{"0":"得到列中前n个最大值对应的索引"}}},"align":"","folded":false,"seq":"auto"}},"doxcn3tXOdoc9TlmGksJB9acOIf":{"id":"doxcn3tXOdoc9TlmGksJB9acOIf","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|e+7e*0+m"},"text":{"0":"df = pd.DataFrame(np.random.randint(1, 15, 15).reshape(5,-1), columns=list('abc'))\nprint(df)\n取'a'列前3个最大值对应的行\nn = 5\ndf['a'].argsort()[::-1].iloc[:3]\n#> a b c\n0 5 5 2\n1 12 7 1\n2 5 2 12\n3 5 14 12\n4 1 13 13\n#> 4 1\n3 3\n2 2\n Name: a, dtype: int64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnnQ4eJoDoAsl5srDrBjt6Bh":{"id":"doxcnnQ4eJoDoAsl5srDrBjt6Bh","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+q"},"text":{"0":"获得dataframe行的和大于100的最末n行索引"}}},"align":"","folded":false,"seq":"auto"}},"doxcnz3VnWQRk4dzXZiHWlLnMBe":{"id":"doxcnz3VnWQRk4dzXZiHWlLnMBe","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|d+9d*0+10"},"text":{"0":"df = pd.DataFrame(np.random.randint(10, 40, 16).reshape(-1, 4))\nprint(df)\ndataframe每行的和\nrowsums = df.apply(np.sum, axis=1)\n选取大于100的最末两行索引\nlast_two_rows = df.iloc[np.where(rowsums > 100)[0][-2:], :]\nnline = np.where(rowsums > 100)[0][-2:]\nnline\n#> 0 1 2 3\n0 19 34 15 12\n1 38 35 14 26\n2 39 32 18 20\n3 28 27 36 38\n#> array([2, 3], dtype=int64)"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnzVIzIjKnTMYPdxdahvwxXe":{"id":"doxcnzVIzIjKnTMYPdxdahvwxXe","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"从series中查找异常值并赋值"}}},"align":"","folded":false,"seq":"auto"}},"doxcnTc79bVTIsQNRpvcu17hQTb":{"id":"doxcnTc79bVTIsQNRpvcu17hQTb","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|9+9t*0+23"},"text":{"0":"ser = pd.Series(np.logspace(-2, 2, 30))\n小于low_per分位的数赋值为low，大于low_per分位的数赋值为high\ndef cap_outliers(ser, low_perc, high_perc):\n low, high = ser.quantile([low_perc, high_perc])\n print(low_perc, '%ile: ', low, '|', high_perc, '%ile: ', high)\n ser[ser < low] = low\n ser[ser > high] = high\n return(ser)\ncapped_ser = cap_outliers(ser, .05, .95)\n#> 0.05 %ile: 0.016049294076965887 | 0.95 %ile: 63.876672220183934"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnkFzCg0nyVHixpMLMZFFiZb":{"id":"doxcnkFzCg0nyVHixpMLMZFFiZb","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+e"},"text":{"0":"交换dataframe的两行"}}},"align":"","folded":false,"seq":"auto"}},"doxcn1yrdo14k0G4lGmq4aim2td":{"id":"doxcn1yrdo14k0G4lGmq4aim2td","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|h+97*0+a"},"text":{"0":"df = pd.DataFrame(np.arange(9).reshape(3, -1))\nprint(df)\n函数\ndef swap_rows(df, i1, i2):\n a, b = df.iloc[i1, :].copy(), df.iloc[i2, :].copy()\n # 通过iloc换行\n df.iloc[i1, :], df.iloc[i2, :] = b, a\n return df\n2和3行互换\nprint(swap_rows(df, 1, 2))\n#> 0 1 2\n0 0 1 2\n1 3 4 5\n2 6 7 8\n#> 0 1 2\n0 0 1 2\n1 6 7 8\n2 3 4 5"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcniETx4so2MIUC95RAV4d5Ne":{"id":"doxcniETx4so2MIUC95RAV4d5Ne","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+d"},"text":{"0":"倒转dataframe的行"}}},"align":"","folded":false,"seq":"auto"}},"doxcnLS29vfQuq9zzuNZuBMwH2f":{"id":"doxcnLS29vfQuq9zzuNZuBMwH2f","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|d+5k*0+a"},"text":{"0":"df = pd.DataFrame(np.arange(9).reshape(3, -1))\nprint(df)\n方法 1\ndf.iloc[::-1, :]\n方法 2\nprint(df.loc[df.index[::-1], :])\n#> 0 1 2\n0 0 1 2\n1 3 4 5\n2 6 7 8\n#> 0 1 2\n2 6 7 8\n1 3 4 5\n0 0 1 2"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnSghN9SmES9VP7zIY3jPZuf":{"id":"doxcnSghN9SmES9VP7zIY3jPZuf","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"对分类变量进行one-hot编码"}}},"align":"","folded":false,"seq":"auto"}},"doxcnf0CSpGm5Mju1ObYqWw0zdd":{"id":"doxcnf0CSpGm5Mju1ObYqWw0zdd","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|g+e2*0+z"},"text":{"0":"df = pd.DataFrame(np.arange(25).reshape(5,-1), columns=list('abcde'))\nprint(df)\n对列'a'进行onehot编码\ndf_onehot = pd.concat([pd.get_dummies(df['a']), df[list('bcde')]], axis=1)\nprint(df_onehot)\n#> a b c d e\n0 0 1 2 3 4\n1 5 6 7 8 9\n2 10 11 12 13 14\n3 15 16 17 18 19\n4 20 21 22 23 24\n#> 0 5 10 15 20 b c d e\n0 1 0 0 0 0 1 2 3 4\n1 0 1 0 0 0 6 7 8 9\n2 0 0 1 0 0 11 12 13 14\n3 0 0 0 1 0 16 17 18 19\n4 0 0 0 0 1 21 22 23 24"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn51aD1EfrMJHjREeN5C0n9d":{"id":"doxcn51aD1EfrMJHjREeN5C0n9d","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+o"},"text":{"0":"获取dataframe行方向上最大值个数最多的列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnx9t7ebCoQUBzzIiyCSedec":{"id":"doxcnx9t7ebCoQUBzzIiyCSedec","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|g+ah*0+17"},"text":{"0":"df = pd.DataFrame(np.random.randint(1,100, 9).reshape(3, -1))\nprint(df)\n获取每列包含行方向上最大值的个数\ncount_series = df.apply(np.argmax, axis=1).value_counts()\nprint(count_series)\n输出行方向最大值个数最多的列的索引\nprint('Column with highest row maxes: ', count_series.index[0])\n#> 0 1 2\n0 46 31 34\n1 38 13 6\n2 1 18 15\n#>统计列的最大值的个数\n 0 2\n 1 1\n dtype: int64\n \n#> Column with highest row maxes: 0"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnZaHdHgV6iI3sLYeXNHSbxX":{"id":"doxcnZaHdHgV6iI3sLYeXNHSbxX","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+c"},"text":{"0":"得到列之间最大的相关系数"}}},"align":"","folded":false,"seq":"auto"}},"doxcnhcuiWHOZvQuBrTzdcJ23xd":{"id":"doxcnhcuiWHOZvQuBrTzdcJ23xd","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|k+ik*0+26"},"text":{"0":"df = pd.DataFrame(np.random.randint(1,100, 16).reshape(4, -1), columns=list('pqrs'), index=list('abcd'))\ndf\nprint(df)\n得到四个列的相关系数\nabs_corrmat = np.abs(df.corr())\nprint(abs_corrmat)\n得到每个列名与其他列的最大相关系数\nmax_corr = abs_corrmat.apply(lambda x: sorted(x)[-2])\n显示每列与其他列的相关系数\nprint('Maximum Correlation possible for each column: ', np.round(max_corr.tolist(), 2))\n#> p q r s\n a 59 99 1 34\n b 89 60 97 40\n c 43 35 14 6\n d 70 59 30 53\n#> p q r s\n p 1.000000 0.200375 0.860051 0.744529\n q 0.200375 1.000000 0.236619 0.438541\n r 0.860051 0.236619 1.000000 0.341399\n s 0.744529 0.438541 0.341399 1.000000\n#> Maximum Correlation possible for each column: [0.86 0.44 0.86 0.74]"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnV32UwAmi3SzZO7yGRpFqMe":{"id":"doxcnV32UwAmi3SzZO7yGRpFqMe","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+h"},"text":{"0":"创建包含每行最小值与最大值比例的列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnFNyc5aJZOtf3WDKIcdO6ag":{"id":"doxcnFNyc5aJZOtf3WDKIcdO6ag","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|f+95*0+f"},"text":{"0":"df = pd.DataFrame(np.random.randint(1,100, 9).reshape(3, -1))\nprint(df)\n方法1：axis=1表示行方向，\nmin_by_max = df.apply(lambda x: np.min(x)/np.max(x), axis=1)\n方法2\nmin_by_max = np.min(df, axis=1)/np.max(df, axis=1)\nmin_by_max\n#> 0 1 2\n0 81 68 59\n1 45 73 23\n2 20 22 69\n \n#> 0 0.728395\n1 0.315068\n2 0.289855\n dtype: float64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnzhDLUJonvruY7hLoynwCne":{"id":"doxcnzhDLUJonvruY7hLoynwCne","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+c"},"text":{"0":"创建包含每行第二大值的列"}}},"align":"","folded":false,"seq":"auto"}},"doxcncY1U0qcolie2HAklxGCsqc":{"id":"doxcncY1U0qcolie2HAklxGCsqc","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|e+9m*0+q"},"text":{"0":"df = pd.DataFrame(np.random.randint(1,100, 9).reshape(3, -1))\nprint(df)\n行方向上取第二大的值组成series\nout = df.apply(lambda x: x.sort_values().unique()[-2], axis=1)\n构建dataframe新的列\ndf['penultimate'] = out\nprint(df)\n#> 0 1 2\n0 28 77 1\n1 43 19 69\n2 29 30 72\n#> 0 1 2 penultimate\n0 28 77 1 28\n1 43 19 69 43\n2 29 30 72 30"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnJWEHUctpQ2T4g8XMg9joig":{"id":"doxcnJWEHUctpQ2T4g8XMg9joig","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"归一化dataframe的所有列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnYNaxHeLQulZ5000m9ur2eb":{"id":"doxcnYNaxHeLQulZ5000m9ur2eb","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|6+6m*0+s"},"text":{"0":"df = pd.DataFrame(np.random.randint(1,100, 80).reshape(8, -1))\n正态分布归一化\nout1 = df.apply(lambda x: ((x - x.mean())/x.std()).round(2))\nprint('Solution Q1\\n',out1)\n线性归一化\nout2 = df.apply(lambda x: ((x.max() - x)/(x.max() - x.min())).round(2))\nprint('Solution Q2\\n', out2)"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnAiYwkWOuhsYF0cGyGmx30b":{"id":"doxcnAiYwkWOuhsYF0cGyGmx30b","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+d"},"text":{"0":"计算每一行与下一行的相关性"}}},"align":"","folded":false,"seq":"auto"}},"doxcn4aKHUakw1uNbVxOQJ9lVtd":{"id":"doxcn4aKHUakw1uNbVxOQJ9lVtd","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|2+21*0+21"},"text":{"0":"df = pd.DataFrame(np.random.randint(1,100, 25).reshape(5, -1))\n行与行之间的相关性\n[df.iloc[i].corr(df.iloc[i+1]).round(2) for i in range(df.shape[0])[:-1]]"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn2zccbsPwbbeikTz7ivprue":{"id":"doxcn2zccbsPwbbeikTz7ivprue","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+n"},"text":{"0":"用0赋值dataframe的主对角线和副对角线"}}},"align":"","folded":false,"seq":"auto"}},"doxcn2lk4mKaTFVKL4CNffpeFhe":{"id":"doxcn2lk4mKaTFVKL4CNffpeFhe","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|i+bo*0+l"},"text":{"0":"df = pd.DataFrame(np.random.randint(1,100, 25).reshape(5, -1))\nprint(df)\nzhu\nfor i in range(df.shape[0]):\n df.iat[i, i] = 0\n df.iat[df.shape[0]-i-1, i] = 0\nprint(df)\n#> 0 1 2 3 4\n0 51 35 71 71 79\n1 78 25 71 85 44\n2 90 97 72 14 4\n3 27 91 37 25 48\n4 1 26 68 70 20\n#> 0 1 2 3 4\n0 0 35 71 71 0\n1 78 0 71 0 44\n2 90 97 0 14 4\n3 27 0 37 0 48\n4 0 26 68 70 0"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn6q3QgVoYARkelwAlKNUyEd":{"id":"doxcn6q3QgVoYARkelwAlKNUyEd","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+o"},"text":{"0":"得到按列分组的dataframe的平均值和标准差"}}},"align":"","folded":false,"seq":"auto"}},"doxcng6v6RRGQkyLaTxlRFrhSSh":{"id":"doxcng6v6RRGQkyLaTxlRFrhSSh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|s+ie*0+j"},"text":{"0":"df = pd.DataFrame({'col1': ['apple', 'banana', 'orange'] * 2,\n 'col2': np.random.randint(0,15,6),\n 'col3': np.random.randint(0, 15, 6)})\nprint(df)\n按列col1分组后的平均值\ndf_grouped_mean = df.groupby(['col1']).mean()\nprint(df_grouped_mean)\n按列col1分组后的标准差\ndf_grouped_std = df.groupby(['col1']).mean()\nprint(df_grouped_std)\n#> col1 col2 col3\n0 apple 2 14\n1 banana 11 8\n2 orange 8 10\n3 apple 5 2\n4 banana 6 12\n5 orange 11 13\n#> col2 col3\n col1\n\n apple 3.5 8.0\n banana 8.5 10.0\n orange 9.5 11.5\n#> col2 col3\n col1\n\n apple 3.5 8.0\n banana 8.5 10.0\n orange 9.5 11.5"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn4DjND4ABdn3H1ypFVXyWHe":{"id":"doxcn4DjND4ABdn3H1ypFVXyWHe","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"得到按列分组后另一列的第n大的值"}}},"align":"","folded":false,"seq":"auto"}},"doxcnr0ukLLEtjJqbzfSvvt4yae":{"id":"doxcnr0ukLLEtjJqbzfSvvt4yae","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|i+ei*0+s"},"text":{"0":"df = pd.DataFrame({'fruit': ['apple', 'banana', 'orange'] * 2,\n 'taste': np.random.rand(6),\n 'price': np.random.randint(0, 15, 6)})\nprint(df)\nteste列按fruit分组\ndf_grpd = df['taste'].groupby(df.fruit)\nteste列中banana元素的信息\nx=df_grpd.get_group('banana')\n排序并找第2大的值\ns = x.sort_values().iloc[-2]\nprint(s)\n#> fruit taste price\n0 apple 0.521990 7\n1 banana 0.640444 0\n2 orange 0.460509 9\n3 apple 0.818963 4\n4 banana 0.646138 7\n5 orange 0.917056 12\n#> 0.6404436436085967"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnBQsMUOgNwfkA8RrVBIGyMb":{"id":"doxcnBQsMUOgNwfkA8RrVBIGyMb","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+t"},"text":{"0":"计算分组dataframe的平均值，并将分组列保留为另一列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnXwQLB8TDLZ0LCCv7JnCODh":{"id":"doxcnXwQLB8TDLZ0LCCv7JnCODh","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|9+8y*0+g"},"text":{"0":"df = pd.DataFrame({'fruit': ['apple', 'banana', 'orange'] * 2,\n 'rating': np.random.rand(6),\n 'price': np.random.randint(0, 15, 6)})\n按fruit分组后，price列的平均值，并将分组置为一列\nout = df.groupby('fruit', as_index=False)['price'].mean()\nprint(out)\n#> fruit price\n0 apple 4.0\n1 banana 6.5\n2 orange 11.0"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnBbNw8Cgyeblldc4zPlV7kf":{"id":"doxcnBbNw8Cgyeblldc4zPlV7kf","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+i"},"text":{"0":"获取两列值元素相等的位置（并非索引）"}}},"align":"","folded":false,"seq":"auto"}},"doxcnHnBZhRsTX4tVnLYMVo9xEf":{"id":"doxcnHnBZhRsTX4tVnLYMVo9xEf","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|9+85*0+10"},"text":{"0":"df = pd.DataFrame({'fruit1': np.random.choice(['apple', 'orange', 'banana'], 3),\n 'fruit2': np.random.choice(['apple', 'orange', 'banana'], 3)})\nprint(df)\n获取两列元素相等的行\nnp.where(df.fruit1 == df.fruit2)\n#> fruit1 fruit2\n0 apple banana\n1 apple apple\n2 orange apple\n#> (array([1], dtype=int64),)"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnf7afRS14zoWtSEBv4eLSGb":{"id":"doxcnf7afRS14zoWtSEBv4eLSGb","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+b"},"text":{"0":"创建指定列偏移后的新列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnQ1CPaaGmu2iwQPnTkHILDe":{"id":"doxcnQ1CPaaGmu2iwQPnTkHILDe","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|b+a2*0+y"},"text":{"0":"df = pd.DataFrame(np.random.randint(1, 100, 20).reshape(-1, 4), columns = list('abcd'))\n创建往下偏移后的列\ndf['a_lag1'] = df['a'].shift(1)\n创建往上偏移后的列\ndf['b_lead1'] = df['b'].shift(-1)\nprint(df)\n#> a b c d a_lag1 b_lead1\n0 29 90 43 24 NaN 36.0\n1 94 36 67 66 29.0 76.0\n2 81 76 44 49 94.0 97.0\n3 55 97 10 74 81.0 43.0\n4 32 43 62 62 55.0 NaN"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcncFdNvImByGYOdYfjCDXi0d":{"id":"doxcncFdNvImByGYOdYfjCDXi0d","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+i"},"text":{"0":"获得dataframe中单一值的频数"}}},"align":"","folded":false,"seq":"auto"}},"doxcnzTzTfnv3MhRd0TH17SwUPR":{"id":"doxcnzTzTfnv3MhRd0TH17SwUPR","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|c+5o*0+d"},"text":{"0":"df = pd.DataFrame(np.random.randint(1, 10, 20).reshape(-1, 4), columns = list('abcd'))\n统计元素值的个数\npd.value_counts(df.values.ravel())\n#> 9 3\n7 3\n3 3\n1 3\n6 2\n5 2\n4 2\n8 1\n2 1\n dtype: int64"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnaw1PhYGBXWbCLv3VY7rs6T":{"id":"doxcnaw1PhYGBXWbCLv3VY7rs6T","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+c"},"text":{"0":"将文本拆分为两个单独的列"}}},"align":"","folded":false,"seq":"auto"}},"doxcnZWjVtLt4gS0B4FSnmJMZph":{"id":"doxcnZWjVtLt4gS0B4FSnmJMZph","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|o+ir*0+u"},"text":{"0":"df = pd.DataFrame([\"STD, City State\",\n\"33, Kolkata West Bengal\",\n\"44, Chennai Tamil Nadu\",\n\"40, Hyderabad Telengana\",\n\"80, Bangalore Karnataka\"], columns=['row'])\nprint(df)\nexpand=True表示以分割符把字符串分成两列\ndf_out = df.row.str.split(',|\\t', expand=True)\n获取新的列\nnew_header = df_out.iloc[0]\n重新赋值\ndf_out = df_out[1:]\ndf_out.columns = new_header\nprint(df_out)\n#> row\n0 STD, City State\n1 33, Kolkata West Bengal\n2 44, Chennai Tamil Nadu\n3 40, Hyderabad Telengana\n4 80, Bangalore Karnataka\n#> 0 STD City State\n1 33 Kolkata West Bengal\n2 44 Chennai Tamil Nadu\n3 40 Hyderabad Telengana\n4 80 Bangalore Karnataka"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcn0bdK9loOZrX71OZS8TnaZe":{"id":"doxcn0bdK9loOZrX71OZS8TnaZe","snapshot":{"type":"ordered","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+g"},"text":{"0":"构建多级索引的dataframe"}}},"align":"","folded":false,"seq":"auto"}},"doxcnNFzejZK2iO8u9hPFpenlee":{"id":"doxcnNFzejZK2iO8u9hPFpenlee","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+w"},"text":{"0":"利用元组（Tuple）构建多级索引，然后定义dataframe."}}},"align":"","folded":false}},"doxcnWadzWtu6aMAkYlNiuoBXJe":{"id":"doxcnWadzWtu6aMAkYlNiuoBXJe","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|c+8v*0+2"},"text":{"0":"如何构建多级索引的dataframe\n先通过元组方式构建多级索引\nimport numpy as np\noutside = ['A','A','A','B','B','B']\ninside =[1,2,3,1,2,3]\nmy_index = list(zip(outside,inside))\nmy_index\n转化为pd格式的索引\nmy_index = pd.MultiIndex.from_tuples(my_index)\nmy_index\n构建多级索引dataframe\ndf = pd.DataFrame(np.random.randn(6,2),index =my_index,columns=['fea1','fea2'])\ndf"}}},"align":"","folded":false,"language":"Python","wrap":false}},"doxcnzuq52tPENUu84BzVDvJPUb":{"id":"doxcnzuq52tPENUu84BzVDvJPUb","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+f"},"text":{"0":"多索引dataframe结果："}}},"align":"","folded":false}},"doxcnE2AK7cK7kIyfD8Az3DkPJc":{"id":"doxcnE2AK7cK7kIyfD8Az3DkPJc","snapshot":{"type":"text","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0+i"},"text":{"0":"获取多索引dataframe的数据："}}},"align":"","folded":false}},"doxcnwOF3CfqlGXRtl46BoyoyOf":{"id":"doxcnwOF3CfqlGXRtl46BoyoyOf","snapshot":{"type":"code","parent_id":"GinpdGIbDoPVO8xUdM1cCNbjnEg","comments":[],"revisions":[],"locked":false,"hidden":false,"author":"7434804997847990275","children":[],"text":{"apool":{"nextNum":1,"numToAttrib":{"0":["author","7434804997847990275"]}},"initialAttributedTexts":{"attribs":{"0":"*0|5+3a*0+t"},"text":{"0":"df.loc['A'].iloc[1]\n#> fea1 -0.794461\n fea2 0.882104\n Name: 2, dtype: float64\ndf.loc['A'].iloc[1]['fea1']\n#> -0.7944609970323794"}}},"align":"","language":"Python","wrap":false}}},"payloadMap":{"doxcnclTVKNrpLZeIe492dVq0Qh":{"level":1},"doxcn2dq7YMhAjCArg6J5VVansb":{"level":1},"doxcn52IGkPyyOjWX04Pt1DPG9d":{"level":1},"doxcncc98qTlXqr0d85X2s8eu8f":{"level":1},"doxcnHZC4BVE2Ji80SWstSfSigg":{"level":1},"doxcnswLs24vvG3JP16ZiXklNre":{"level":1},"doxcnzNadbox0muCzQStBlkWFBd":{"level":1},"doxcnxNO5DBjsKA3hHXr4ZdsjDb":{"level":1},"doxcnjhca9AVbknM7gRhzPc43we":{"level":1},"doxcnBOt3E2IPjglm9bgS3ARGte":{"level":1},"doxcnNFzejZK2iO8u9hPFpenlee":{"level":1},"doxcnzuq52tPENUu84BzVDvJPUb":{"level":1},"doxcnE2AK7cK7kIyfD8Az3DkPJc":{"level":1}},"extra":{"channel":"saas","pasteRandomId":"3258a30c-84f4-40fb-b39e-651741c01c1e","mention_page_title":{},"external_mention_url":{},"isEqualBlockSelection":true},"isKeepQuoteContainer":false,"selection":[{"id":1,"type":"text","selection":{"start":0,"end":10},"recordId":"GinpdGIbDoPVO8xUdM1cCNbjnEg"},{"id":2,"type":"text","selection":{"start":0,"end":7},"recordId":"doxcnclTVKNrpLZeIe492dVq0Qh"},{"id":3,"type":"text","selection":{"start":0,"end":10},"recordId":"doxcnay0d7qZo8tbQ7FBF9FWtUd"},{"id":4,"type":"text","selection":{"start":0,"end":35},"recordId":"doxcn2dq7YMhAjCArg6J5VVansb"},{"id":5,"type":"block","recordId":"doxcndTngkOURHCdoWrwaRc8xqf"},{"id":6,"type":"text","selection":{"start":0,"end":36},"recordId":"doxcn52IGkPyyOjWX04Pt1DPG9d"},{"id":7,"type":"block","recordId":"doxcniY8Y3QJAW5DRdDFzT8GZqd"},{"id":8,"type":"text","selection":{"start":0,"end":21},"recordId":"doxcnd9ntFL09hjl66lgtcM2nEc"},{"id":9,"type":"block","recordId":"doxcnzZtiflNw5vNLvwvN8lfAhg"},{"id":10,"type":"text","selection":{"start":0,"end":12},"recordId":"doxcncLmWdikkZkDCE3Ka5AMglc"},{"id":11,"type":"text","selection":{"start":0,"end":32},"recordId":"doxcncc98qTlXqr0d85X2s8eu8f"},{"id":12,"type":"text","selection":{"start":0,"end":65},"recordId":"doxcnHZC4BVE2Ji80SWstSfSigg"},{"id":13,"type":"block","recordId":"doxcn1U5V30YIrU61rpRrBRTbXd"},{"id":14,"type":"text","selection":{"start":0,"end":61},"recordId":"doxcnswLs24vvG3JP16ZiXklNre"},{"id":15,"type":"block","recordId":"doxcnXAof7jWLxGZknAZa1bkCih"},{"id":16,"type":"text","selection":{"start":0,"end":11},"recordId":"doxcnMju63hRn8ngF31l6Fb5Vxh"},{"id":17,"type":"text","selection":{"start":0,"end":17},"recordId":"doxcnIZbax2lqCv0wTAiOxgWEfe"},{"id":18,"type":"block","recordId":"doxcnpAsutfNcnQXrvFvFaDUuIf"},{"id":19,"type":"text","selection":{"start":0,"end":25},"recordId":"doxcnTRHNLSwjSx7FysZ1tNCA4e"},{"id":20,"type":"block","recordId":"doxcnzgvPKzztjypBge56vtVkZf"},{"id":21,"type":"text","selection":{"start":0,"end":21},"recordId":"doxcnQWDbvCUgAb7TNKatSeejee"},{"id":22,"type":"block","recordId":"doxcnTiOGA2zHaksZ5uOCgMxofe"},{"id":23,"type":"text","selection":{"start":0,"end":8},"recordId":"doxcntVGXD5ZeGYw179Mca8TPQb"},{"id":24,"type":"block","recordId":"doxcn8frvSrzlSlTVh9ac3uG9Bb"},{"id":25,"type":"text","selection":{"start":0,"end":27},"recordId":"doxcnufiEauX5gsP79cGhaUYA6c"},{"id":26,"type":"block","recordId":"doxcnjScfKt3D5AzKCQxvwKWDUb"},{"id":27,"type":"text","selection":{"start":0,"end":22},"recordId":"doxcnoRhxOpX2v7SMai3zl8Ivvd"},{"id":28,"type":"block","recordId":"doxcnCMV0TGhZ6G4RyyEj73j44f"},{"id":29,"type":"text","selection":{"start":0,"end":15},"recordId":"doxcnM6WCByb3uUqio3AYsdPWyf"},{"id":30,"type":"block","recordId":"doxcnUANTw2hmXue6QqizEizWKh"},{"id":31,"type":"text","selection":{"start":0,"end":17},"recordId":"doxcnsNOyxbu7zkVsNlATUZt0ub"},{"id":32,"type":"block","recordId":"doxcnIdLKJwKuNEv698aFfXNlIt"},{"id":33,"type":"text","selection":{"start":0,"end":32},"recordId":"doxcn2K7O1vuEhkG1DO4B6Ddm6e"},{"id":34,"type":"block","recordId":"doxcn6ZTmzwk6veIzfDktgd0Ssh"},{"id":35,"type":"text","selection":{"start":0,"end":20},"recordId":"doxcnSc7yZ2BuuIYzM8pT5KZviw"},{"id":36,"type":"text","selection":{"start":0,"end":39},"recordId":"doxcnzNadbox0muCzQStBlkWFBd"},{"id":37,"type":"block","recordId":"doxcnyZOFv3gGqZBlNVMe9UqOih"},{"id":38,"type":"text","selection":{"start":0,"end":25},"recordId":"doxcnaI5I9koRMU2sa6FOXP67Ad"},{"id":39,"type":"block","recordId":"doxcn5621kGVGfpskfAUSrh1Vjd"},{"id":40,"type":"text","selection":{"start":0,"end":18},"recordId":"doxcnhu5fOuL30Y4TPy7GTLj6Bg"},{"id":41,"type":"block","recordId":"doxcnDi7S0HztqExvkWEsgmZVqh"},{"id":42,"type":"text","selection":{"start":0,"end":23},"recordId":"doxcnjZK8AZIRC0m3fcEKIaHWTb"},{"id":43,"type":"block","recordId":"doxcnIbwFpZB2Zucx30Pzbfo27c"},{"id":44,"type":"block","recordId":"doxcn4Hl615SXgJOfgej9h3QQ0f"},{"id":45,"type":"block","recordId":"doxcnK5gaYyuBZrfEYZlOFTKeJh"},{"id":46,"type":"text","selection":{"start":0,"end":28},"recordId":"doxcnJXasa1u6tB8EuNdHHJ6emg"},{"id":47,"type":"block","recordId":"doxcnnykCiu3OrIZOLvUqLh3wKg"},{"id":48,"type":"text","selection":{"start":0,"end":14},"recordId":"doxcn6CEeMJyPy9K9NnA5ryfTWg"},{"id":49,"type":"block","recordId":"doxcn0ML0hiLMnajFcK714dhnid"},{"id":50,"type":"text","selection":{"start":0,"end":19},"recordId":"doxcndp7mEwLbDg7dYegIkjWdnd"},{"id":51,"type":"block","recordId":"doxcnZnzRDZS1hwIE5GJ8MMp8eg"},{"id":52,"type":"text","selection":{"start":0,"end":19},"recordId":"doxcnwJPX062FrA8Gw15LXfcyrc"},{"id":53,"type":"block","recordId":"doxcn4dC69t4oNWCx3dPI5FQlqf"},{"id":54,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcngSrF1MHJESrXvsLXcmyAVa"},{"id":55,"type":"block","recordId":"doxcn56NDlN9nww0ZawWiSxfkpf"},{"id":56,"type":"text","selection":{"start":0,"end":22},"recordId":"doxcnmnvJtVb5xifEJZMP28ycOd"},{"id":57,"type":"block","recordId":"doxcnbb2yY0wCcdRTYOfZVp4zlh"},{"id":58,"type":"text","selection":{"start":0,"end":23},"recordId":"doxcnpIgwqUowns3ly5JhhybMDd"},{"id":59,"type":"block","recordId":"doxcnkBpkcWb4h5aeSoRmeZZ2ub"},{"id":60,"type":"text","selection":{"start":0,"end":25},"recordId":"doxcnn5kOt9nLdqzpKJhRO1NEsk"},{"id":61,"type":"block","recordId":"doxcnXYaC5zCdivW7gtNo0IG8yb"},{"id":62,"type":"text","selection":{"start":0,"end":17},"recordId":"doxcnMN7OrEgBd5p3VY542KBUBe"},{"id":63,"type":"block","recordId":"doxcnyG10bN7OcZetGXaMC2Ssge"},{"id":64,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcnE3JY4UP13lsWX2YAV1UEKb"},{"id":65,"type":"text","selection":{"start":0,"end":15},"recordId":"doxcnxNO5DBjsKA3hHXr4ZdsjDb"},{"id":66,"type":"block","recordId":"doxcnvzAvBDNNccupAPpapRoLde"},{"id":67,"type":"text","selection":{"start":0,"end":13},"recordId":"doxcnpfBtvLEbJb5Ow30wwahi6g"},{"id":68,"type":"block","recordId":"doxcnW96M1lVwQevxGHIdAwsBob"},{"id":69,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcnFe2PlkgVgS0CpMpgLUhKif"},{"id":70,"type":"block","recordId":"doxcnmRLDwpeCuVR4c3XXuP9pMf"},{"id":71,"type":"text","selection":{"start":0,"end":15},"recordId":"doxcnShQpxjmFWcKKj0ZbV8ymQc"},{"id":72,"type":"block","recordId":"doxcnySbPoaoBTkdhopxWtexQBd"},{"id":73,"type":"text","selection":{"start":0,"end":0},"recordId":"doxcnjhca9AVbknM7gRhzPc43we"},{"id":74,"type":"text","selection":{"start":0,"end":15},"recordId":"doxcnBbXCVQeOGUtZmhRQzdBjji"},{"id":75,"type":"block","recordId":"doxcnY6TrasVJsYBNNMeq4KlORS"},{"id":76,"type":"text","selection":{"start":0,"end":23},"recordId":"doxcnZ5Zpn4NoVc3lFU8KMQOlpE"},{"id":77,"type":"block","recordId":"doxcnNLqv4k0rbPpING3fmLw2gg"},{"id":78,"type":"text","selection":{"start":0,"end":12},"recordId":"doxcnVSeSN29rmGxOHh91gXtG2b"},{"id":79,"type":"text","selection":{"start":0,"end":47},"recordId":"doxcnBOt3E2IPjglm9bgS3ARGte"},{"id":80,"type":"block","recordId":"doxcnHVnPDKiU7yv8BTCg9h5qPe"},{"id":81,"type":"text","selection":{"start":0,"end":12},"recordId":"doxcnq88mUIGE4KxyihW5RDPI61"},{"id":82,"type":"block","recordId":"doxcnq7NY1oPLs8GZr7X639vF9e"},{"id":83,"type":"text","selection":{"start":0,"end":32},"recordId":"doxcnos8LJdTZm9JDyQcoS0wyB0"},{"id":84,"type":"block","recordId":"doxcnhQ8ai47rk7YI7XxTlRmcGO"},{"id":85,"type":"text","selection":{"start":0,"end":10},"recordId":"doxcnGr6WT4IBoEy31MrS2YEWKh"},{"id":86,"type":"block","recordId":"doxcnOi7Q28BU1nsulYmcCHOyrd"},{"id":87,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcnADu5qHAmF21NboPXCIqDUf"},{"id":88,"type":"block","recordId":"doxcnsreRpfdnYnVxKfd8XxCvKh"},{"id":89,"type":"text","selection":{"start":0,"end":18},"recordId":"doxcnxHedW5jNMRsSfC02ZAThng"},{"id":90,"type":"block","recordId":"doxcnffYje0A03g0sRrQq8HQtYd"},{"id":91,"type":"text","selection":{"start":0,"end":21},"recordId":"doxcnfT18gUJC1e5lRkiS2xIgDc"},{"id":92,"type":"block","recordId":"doxcniNaO0YF7eHot9uvcOFRmIs"},{"id":93,"type":"text","selection":{"start":0,"end":13},"recordId":"doxcnytYH8AN71cB1KJXR2w6zMg"},{"id":94,"type":"block","recordId":"doxcnDkS4YLb4ejiNvIxDSwJ4e9"},{"id":95,"type":"text","selection":{"start":0,"end":25},"recordId":"doxcn7ETXSUxT9qpIf49tAIGkeg"},{"id":96,"type":"block","recordId":"doxcnJ9wLnv4yuW3WxXdeaCHeUh"},{"id":97,"type":"text","selection":{"start":0,"end":19},"recordId":"doxcn7b3mdmtAABCkuyuGXkYfId"},{"id":98,"type":"block","recordId":"doxcnnTBI5dltwM73MHHNIl4Ibh"},{"id":99,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcnY31bnFKqQazJFWmPQjJtTd"},{"id":100,"type":"block","recordId":"doxcnzPg7cUvsARaTQDnAGTTloc"},{"id":101,"type":"text","selection":{"start":0,"end":14},"recordId":"doxcnXBBBAM1JPSTzPt22lPBkZc"},{"id":102,"type":"block","recordId":"doxcnzmi8YLHvbV1Sm8HFaxudJf"},{"id":103,"type":"text","selection":{"start":0,"end":24},"recordId":"doxcnEe0kvQVwnd7bHsQ1DkQLUe"},{"id":104,"type":"block","recordId":"doxcn5A91Y5O3Nt3kUBML0RQVwC"},{"id":105,"type":"text","selection":{"start":0,"end":26},"recordId":"doxcnlVWqEr1twoRfj4wVLeURXb"},{"id":106,"type":"block","recordId":"doxcnTDMsRcNOikYW9m1S6nfsDu"},{"id":107,"type":"text","selection":{"start":0,"end":15},"recordId":"doxcncPj13pCOHRnrFuiFRQr7cc"},{"id":108,"type":"block","recordId":"doxcn3tXOdoc9TlmGksJB9acOIf"},{"id":109,"type":"text","selection":{"start":0,"end":26},"recordId":"doxcnnQ4eJoDoAsl5srDrBjt6Bh"},{"id":110,"type":"block","recordId":"doxcnz3VnWQRk4dzXZiHWlLnMBe"},{"id":111,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcnzVIzIjKnTMYPdxdahvwxXe"},{"id":112,"type":"block","recordId":"doxcnTc79bVTIsQNRpvcu17hQTb"},{"id":113,"type":"text","selection":{"start":0,"end":14},"recordId":"doxcnkFzCg0nyVHixpMLMZFFiZb"},{"id":114,"type":"block","recordId":"doxcn1yrdo14k0G4lGmq4aim2td"},{"id":115,"type":"text","selection":{"start":0,"end":13},"recordId":"doxcniETx4so2MIUC95RAV4d5Ne"},{"id":116,"type":"block","recordId":"doxcnLS29vfQuq9zzuNZuBMwH2f"},{"id":117,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcnSghN9SmES9VP7zIY3jPZuf"},{"id":118,"type":"block","recordId":"doxcnf0CSpGm5Mju1ObYqWw0zdd"},{"id":119,"type":"text","selection":{"start":0,"end":24},"recordId":"doxcn51aD1EfrMJHjREeN5C0n9d"},{"id":120,"type":"block","recordId":"doxcnx9t7ebCoQUBzzIiyCSedec"},{"id":121,"type":"text","selection":{"start":0,"end":12},"recordId":"doxcnZaHdHgV6iI3sLYeXNHSbxX"},{"id":122,"type":"block","recordId":"doxcnhcuiWHOZvQuBrTzdcJ23xd"},{"id":123,"type":"text","selection":{"start":0,"end":17},"recordId":"doxcnV32UwAmi3SzZO7yGRpFqMe"},{"id":124,"type":"block","recordId":"doxcnFNyc5aJZOtf3WDKIcdO6ag"},{"id":125,"type":"text","selection":{"start":0,"end":12},"recordId":"doxcnzhDLUJonvruY7hLoynwCne"},{"id":126,"type":"block","recordId":"doxcncY1U0qcolie2HAklxGCsqc"},{"id":127,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcnJWEHUctpQ2T4g8XMg9joig"},{"id":128,"type":"block","recordId":"doxcnYNaxHeLQulZ5000m9ur2eb"},{"id":129,"type":"text","selection":{"start":0,"end":13},"recordId":"doxcnAiYwkWOuhsYF0cGyGmx30b"},{"id":130,"type":"block","recordId":"doxcn4aKHUakw1uNbVxOQJ9lVtd"},{"id":131,"type":"text","selection":{"start":0,"end":23},"recordId":"doxcn2zccbsPwbbeikTz7ivprue"},{"id":132,"type":"block","recordId":"doxcn2lk4mKaTFVKL4CNffpeFhe"},{"id":133,"type":"text","selection":{"start":0,"end":24},"recordId":"doxcn6q3QgVoYARkelwAlKNUyEd"},{"id":134,"type":"block","recordId":"doxcng6v6RRGQkyLaTxlRFrhSSh"},{"id":135,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcn4DjND4ABdn3H1ypFVXyWHe"},{"id":136,"type":"block","recordId":"doxcnr0ukLLEtjJqbzfSvvt4yae"},{"id":137,"type":"text","selection":{"start":0,"end":29},"recordId":"doxcnBQsMUOgNwfkA8RrVBIGyMb"},{"id":138,"type":"block","recordId":"doxcnXwQLB8TDLZ0LCCv7JnCODh"},{"id":139,"type":"text","selection":{"start":0,"end":18},"recordId":"doxcnBbNw8Cgyeblldc4zPlV7kf"},{"id":140,"type":"block","recordId":"doxcnHnBZhRsTX4tVnLYMVo9xEf"},{"id":141,"type":"text","selection":{"start":0,"end":11},"recordId":"doxcnf7afRS14zoWtSEBv4eLSGb"},{"id":142,"type":"block","recordId":"doxcnQ1CPaaGmu2iwQPnTkHILDe"},{"id":143,"type":"text","selection":{"start":0,"end":18},"recordId":"doxcncFdNvImByGYOdYfjCDXi0d"},{"id":144,"type":"block","recordId":"doxcnzTzTfnv3MhRd0TH17SwUPR"},{"id":145,"type":"text","selection":{"start":0,"end":12},"recordId":"doxcnaw1PhYGBXWbCLv3VY7rs6T"},{"id":146,"type":"block","recordId":"doxcnZWjVtLt4gS0B4FSnmJMZph"},{"id":147,"type":"text","selection":{"start":0,"end":16},"recordId":"doxcn0bdK9loOZrX71OZS8TnaZe"},{"id":148,"type":"text","selection":{"start":0,"end":32},"recordId":"doxcnNFzejZK2iO8u9hPFpenlee"},{"id":149,"type":"block","recordId":"doxcnWadzWtu6aMAkYlNiuoBXJe"},{"id":150,"type":"text","selection":{"start":0,"end":15},"recordId":"doxcnzuq52tPENUu84BzVDvJPUb"},{"id":151,"type":"text","selection":{"start":0,"end":18},"recordId":"doxcnE2AK7cK7kIyfD8Az3DkPJc"},{"id":152,"type":"block","recordId":"doxcnwOF3CfqlGXRtl46BoyoyOf"}],"pasteFlag":"789fc78c-a1b7-4ecd-a121-33ddc6b9c382"}" data-lark-record-format="docx/record" class="lark-record-clipboard">

最全总结！Pandas 十大清洗技巧 （下）

最全总结！Pandas 十大清洗技巧（下）