6626070
2997924

PL03-Topic02, Pandas

Back to the previous page
List of posts to read before reading this article

Contents

Installation

For linux

$




For windows




Version Control




Series

One-column

Creating and searching

METHOD 1, Creating Series : sperately

import pandas as pd

s = pd.Series([909976, 8615246, 2872086, 2273305])
s.name = "Population"
s.index = ["Stockholm", "London", "Rome", "Paris"] 

s
OUTPUT 
Stockholm     909976
London       8615246
Rome         2872086
Paris        2273305
Name: Population, dtype: int64



type(s)
OUTPUT 
<class 'pandas.core.series.Series'> 



type(s.name)
OUTPUT 
<class 'str'> 



type(s.index)
OUTPUT 
<class 'pandas.core.indexes.base.Index'>




METHOD2, Creating Series : all at once

import pandas as pd

s = pd.Series([909976, 8615246, 2872086, 2273305], 
              name="Population" ,
              index=["Stockholm", "London", "Rome", "Paris"])

s
OUTPUT 
Stockholm     909976
London       8615246
Rome         2872086
Paris        2273305
Name: Population, dtype: int64



s.index
OUTPUT 
Index(['Stockholm', 'London', 'Rome', 'Paris'], dtype='object')



s.name
OUTPUT 
'Population'



s.values
OUTPUT

method searching all values

array([ 909976, 8615246, 2872086, 2273305], dtype=int64)



s[1]
OUTPUT

method0 : searching single value

8615246



s["London"]
OUTPUT

method1 : searching single value

8615246



s.London
OUTPUT

method2 : searching single value

8615246



s[[1,2]]
OUTPUT

method0 : searching multi-values

London    8615246
Rome      2872086
Name: Population, dtype: int64



s[["London","Rome"]]
OUTPUT

method1 : searching multi-values

London    8615246
Rome      2872086
Name: Population, dtype: int64



s[1:3]
OUTPUT

method2 : searching multi-values

London    8615246
Rome      2872086
Name: Population, dtype: int64



s["London":"Rome"]
OUTPUT

method3 : searching multi-values

London    8615246
Rome      2872086
Name: Population, dtype: int64




Analysis

import pandas as pd

s = pd.Series([1,1,1,2,2,2,2,2,3,3,3,3,3,3,3,3,3,3,4,4,4], 
              name="Population")

s.head()
OUTPUT 
0    1
1    1
2    1
3    2
4    2
Name: Population, dtype: int64



s.shape
OUTPUT 
(21,)



s.unique()
OUTPUT 
array([1, 2, 3, 4], dtype=int64)



s.value_counts()
OUTPUT 
3    10
2     5
4     3
1     3
Name: Population, dtype: int64




Statistics

import pandas as pd
import matplotlib.pyplot as plt

s = pd.Series([1,1,1,2,2,2,2,2,3,3,3,3,3,3,3,3,3,3,4,4,4], 
              name="Population")

s.median(), s.mean(), s.std(), s.min(), s.max()
OUTPUT 
(3.0, 2.619047619047619, 0.9206622874969125, 1, 4)



s.describe()
OUTPUT 
count    21.000000
mean      2.619048
std       0.920662
min       1.000000
25%       2.000000
50%       3.000000
75%       3.000000
max       4.000000
Name: Population, dtype: float64



fig, axes = plt.subplots(1,4, figsize=(12, 3))
s.plot(ax=axes[0], kind='line', title='line')
s.plot(ax=axes[1], kind='bar', title='bar')
s.plot(ax=axes[2], kind='box', title='box')
s.plot(ax=axes[3], kind='pie', title='pie')
OUTPUT

다운로드




Several columns

Creating

Concatenate

import pandas as pd

s1 = pd.Series([909976, 8615246, 2872086, 2273305], 
              name="Population1" ,
              index=["Stockholm1", "London1", "Rome1", "Paris1"])
s2 = pd.Series([909976, 8615246, 2872086, 2273305], 
              name="Population2" ,
              index=["Stockholm2", "London2", "Rome2", "Paris2"])
s3 = pd.Series([909976, 8615246, 2872086, 2273305], 
              name="Population3" ,
              index=["Stockholm3", "London3", "Rome3", "Paris3"])

s1, s2, s3
OUTPUT 
Stockholm1     909976
London1       8615246
Rome1         2872086
Paris1        2273305
Name: Population1, dtype: int64 

 Stockholm2     909976
London2       8615246
Rome2         2872086
Paris2        2273305
Name: Population2, dtype: int64 

 Stockholm3     909976
London3       8615246
Rome3         2872086
Paris3        2273305
Name: Population3, dtype: int64



pd.concat([s1, s2, s3], axis=0)
OUTPUT 
Stockholm1     909976
London1       8615246
Rome1         2872086
Paris1        2273305
Stockholm2     909976
London2       8615246
Rome2         2872086
Paris2        2273305
Stockholm3     909976
London3       8615246
Rome3         2872086
Paris3        2273305
dtype: int64



pd.concat([s1, s2, s3], axis=1)
OUTPUT 
            Population1  Population2  Population3
London1       8615246.0          NaN          NaN
London2             NaN    8615246.0          NaN
London3             NaN          NaN    8615246.0
Paris1        2273305.0          NaN          NaN
Paris2              NaN    2273305.0          NaN
Paris3              NaN          NaN    2273305.0
Rome1         2872086.0          NaN          NaN
Rome2               NaN    2872086.0          NaN
Rome3               NaN          NaN    2872086.0
Stockholm1     909976.0          NaN          NaN
Stockholm2          NaN     909976.0          NaN
Stockholm3          NaN          NaN     909976.0



pd.concat([s1, s2, s3], axis=1, ignore_index=True)
OUTPUT 
                    0          1          2
London1     8615246.0        NaN        NaN
London2           NaN  8615246.0        NaN
London3           NaN        NaN  8615246.0
Paris1      2273305.0        NaN        NaN
Paris2            NaN  2273305.0        NaN
Paris3            NaN        NaN  2273305.0
Rome1       2872086.0        NaN        NaN
Rome2             NaN  2872086.0        NaN
Rome3             NaN        NaN  2872086.0
Stockholm1   909976.0        NaN        NaN
Stockholm2        NaN   909976.0        NaN
Stockholm3        NaN        NaN   909976.0



pd.concat([s1, s2, s3], axis=1, keys=['C0', 'C1', 'C2'])
OUTPUT 
                   C0         C1         C2
London1     8615246.0        NaN        NaN
London2           NaN  8615246.0        NaN
London3           NaN        NaN  8615246.0
Paris1      2273305.0        NaN        NaN
Paris2            NaN  2273305.0        NaN
Paris3            NaN        NaN  2273305.0
Rome1       2872086.0        NaN        NaN
Rome2             NaN  2872086.0        NaN
Rome3             NaN        NaN  2872086.0
Stockholm1   909976.0        NaN        NaN
Stockholm2        NaN   909976.0        NaN
Stockholm3        NaN        NaN   909976.0




get_dummies

import pandas as pd

s1 = pd.Series(list('abca'))
s2 = pd.get_dummies(s1)
OUTPUT 
s1

0    a
1    b
2    c
3    a
dtype: object



s2

	a	b	c
0	1	0	0
1	0	1	0
2	0	0	1
3	1	0	0




DataFrame

One-dataframe

Creating and searching

METHOD 1, Creating frame based on row

import pandas as pd

df = pd.DataFrame([[909976, "Sweden"],
                   [8615246, "United Kingdom"],
                   [2872086, "Italy"],
                   [2273305, "France"]])
df.index = ["Stockholm", "London", "Rome", "Paris"]
df.columns = ["Population", "State"] 

df
OUTPUT 
           Population           State
Stockholm      909976          Sweden
London        8615246  United Kingdom
Rome          2872086           Italy
Paris         2273305          France




METHOD 2, Creating frame based on row

import pandas as pd

df = pd.DataFrame([[909976, "Sweden"],
                   [8615246, "United Kingdom"],
                   [2872086, "Italy"],
                   [2273305, "France"]])
df.index = ["Stockholm", "London", "Rome", "Paris"]
df.rename(columns={0:"Population", 1:"State"}, inplace=True)

df
OUTPUT 
           Population           State
Stockholm      909976          Sweden
London        8615246  United Kingdom
Rome          2872086           Italy
Paris         2273305          France




METHOD 3, Creating frame based on row, all at once

import pandas as pd

df = pd.DataFrame([[909976, "Sweden"],
                   [8615246, "United Kingdom"],
                   [2872086, "Italy"],
                   [2273305, "France"]],
                 index=["Stockholm", "London", "Rome", "Paris"],
                 columns=["Population", "State"])

df
OUTPUT 
           Population           State
Stockholm      909976          Sweden
London        8615246  United Kingdom
Rome          2872086           Italy
Paris         2273305          France




METHOD 1, Creating frame based on columns, all at once

import pandas as pd

df = pd.DataFrame({"Population": [909976, 8615246, 2872086, 2273305],
                   "State": ["Sweden", "United Kingdom", "Italy", "France"]},
                  index=["Stockholm", "London", "Rome", "Paris"])

df
OUTPUT 
           Population           State
Stockholm      909976          Sweden
London        8615246  United Kingdom
Rome          2872086           Italy
Paris         2273305          France



df.index
OUTPUT

searching all index

Index(['Stockholm', 'London', 'Rome', 'Paris'], dtype='object', name='index')



df.index.difference(['Stockholm'])
OUTPUT 
Index(['London', 'Paris', 'Rome'], dtype='object')



df.columns
OUTPUT

searching all columns

Index(['Population', 'State'], dtype='object')



df.columns.difference(['Population'])
OUTPUT 
Index(['State'], dtype='object')


Searching row or values of row

df.loc["Stockholm"]
OUTPUT

method searching single row

Population    909976
State         Sweden
Name: Stockholm, dtype: object



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df.loc["Stockholm"])
OUTPUT : <class 'pandas.core.series.Series'>



df.loc[["Paris","Rome"]]
OUTPUT

method1 searching multi-rows

       Population   State                  
Paris     2273305  France
Rome      2872086   Italy



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df.loc[["Paris","Rome"]])
OUTPUT : <class 'pandas.core.frame.DataFrame'>



df[2:4]
OUTPUT

method2 searching multi-rows

       Population   State                
Rome      2872086   Italy
Paris     2273305  France



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df[2:4])
OUTPUT : <class 'pandas.core.frame.DataFrame'>



df.loc["Stockholm","Population"]
OUTPUT

method1 searching single value

909976



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df.loc["Stockholm","Population"])
OUTPUT : <class 'numpy.int64'>



df.loc["Stockholm"][0]
OUTPUT

method2 searching single value

909976



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df.loc["Stockholm"][0])
OUTPUT : <class 'numpy.int64'>



df.loc["Stockholm"]["Population"]
OUTPUT

method3 searching single value

909976



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df.loc["Stockholm"]["Population"])
OUTPUT : <class 'numpy.int64'>



df.loc["Stockholm"].Population
OUTPUT

method4 searching single value

909976



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df.loc["Stockholm"].Population)
OUTPUT :  <class 'numpy.int64'>



df.loc[["Paris","Rome"],"Population"]
OUTPUT

searching multi-values

Paris    2273305
Rome     2872086
Name: Population, dtype: int64



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df.loc[["Paris","Rome"],"Population"])
OUTPUT : <class 'pandas.core.series.Series'>


Searching columns or values of columns

df['Population']
OUTPUT

method1 searching single column

Stockholm     909976
London       8615246
Rome         2872086
Paris        2273305
Name: Population, dtype: int64



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df['Population'])
OUTPUT : <class 'pandas.core.series.Series'>



df.Population
OUTPUT

method2 searching single column

Stockholm     909976
London       8615246
Rome         2872086
Paris        2273305
Name: Population, dtype: int64



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France


Data-type

 INPUT : type(df.Population)
OUTPUT : <class 'pandas.core.series.Series'>



df['Population'][0]
OUTPUT

method1 searching single value

909976



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France 



df.Population[0]
OUTPUT

method2 searching single value

909976



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France 



df['Population']['Stockholm']
OUTPUT

method3 searching single value

909976



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France 



df['Population'].Stockholm
OUTPUT

method4 searching single value

909976



Original dataset

  Population State
Stockholm 909976 Sweden
London 8615246 United Kingdom
Rome 2872086 Italy
Paris 2273305 France





METHOD 2, Creating frame based on columns

import pandas as pd

df = pd.DataFrame({"Population": [909976, 8615246, 2872086, 2273305],
                   "State": ["Sweden", "United Kingdom", "Italy", "France"],
                   "index": ["Stockholm", "London", "Rome", "Paris"]})
df = df.set_index("index")

df
OUTPUT 
           Population           State
index                                
Stockholm      909976          Sweden
London        8615246  United Kingdom
Rome          2872086           Italy
Paris         2273305          France



df.iloc[0,:]

Population    909976
State         Sweden
Name: Stockholm, dtype: object



df.iloc[:,0]

index
Stockholm     909976
London       8615246
Rome         2872086
Paris        2273305
Name: Population, dtype: int64




Arrangement

STEP1

import pandas as pd

# creating dataset
df = pd.DataFrame({"Population": [909976, 8615246, 2872086, 2273305,123234,123444,23333,343434],
                   "State": ["Sweden", "United Kingdom", "Italy","Seoul","Suwon", "France","Korea", "Japan"],
                   "Alphabet" : ["a","b","x","d","a","a","b","c"],
                   "rank" : [1,2,3,4,6,5,7,8]})
df
OUTPUT 
   Population           State Alphabet  rank
0      909976          Sweden        a     1
1     8615246  United Kingdom        b     2
2     2872086           Italy        x     3
3     2273305           Seoul        d     4
4      123234           Suwon        a     6
5      123444          France        a     5
6       23333           Korea        b     7
7      343434           Japan        c     8


STEP2

df = df.sort_index(axis=1)
df
OUTPUT 
  Alphabet  Population           State  rank
0        a      909976          Sweden     1
1        b     8615246  United Kingdom     2
2        x     2872086           Italy     3
3        d     2273305           Seoul     4
4        a      123234           Suwon     6
5        a      123444          France     5
6        b       23333           Korea     7
7        c      343434           Japan     8


STEP3

df = df.set_index(['Alphabet','rank'])
df
OUTPUT 
               Population           State
Alphabet rank                            
a        1         909976          Sweden
b        2        8615246  United Kingdom
x        3        2872086           Italy
d        4        2273305           Seoul
a        6         123234           Suwon
         5         123444          France
b        7          23333           Korea
c        8         343434           Japan


STEP4

df = df.sort_index()
df
OUTPUT 
               Population           State
Alphabet rank                            
a        1         909976          Sweden
         5         123444          France
         6         123234           Suwon
b        2        8615246  United Kingdom
         7          23333           Korea
c        8         343434           Japan
d        4        2273305           Seoul
x        3        2872086           Italy


Based on rank

df.sort_values("rank", ascending=False)
OUTPUT 
               Population           State
Alphabet rank                            
c        8         343434           Japan
b        7          23333           Korea
a        6         123234           Suwon
         5         123444          France
d        4        2273305           Seoul
x        3        2872086           Italy
b        2        8615246  United Kingdom
a        1         909976          Sweden


Based on Population

df.sort_values("Population", ascending=False)
OUTPUT 
               Population           State
Alphabet rank                            
b        2        8615246  United Kingdom
x        3        2872086           Italy
d        4        2273305           Seoul
a        1         909976          Sweden
c        8         343434           Japan
a        5         123444          France
         6         123234           Suwon
b        7          23333           Korea


Based on State

df.sort_values("State", ascending=False)
OUTPUT 
               Population           State
Alphabet rank                            
b        2        8615246  United Kingdom
a        1         909976          Sweden
         6         123234           Suwon
d        4        2273305           Seoul
b        7          23333           Korea
c        8         343434           Japan
x        3        2872086           Italy
a        5         123444          France


Based on Alphabet

df = df.sort_values("Alphabet", ascending=False)
df
OUTPUT 
               Population           State
Alphabet rank                            
x        3        2872086           Italy
d        4        2273305           Seoul
c        8         343434           Japan
b        2        8615246  United Kingdom
         7          23333           Korea
a        1         909976          Sweden
         6         123234           Suwon
         5         123444          France




Deleting

drop

import pandas as pd
import numpy as np

df = pd.DataFrame(np.arange(12).reshape(3, 4),
                  columns=['A', 'B', 'C', 'D'])
df
OUTPUT 
   A  B   C   D
0  0  1   2   3
1  4  5   6   7
2  8  9  10  11



# single column drop
df.drop('A', axis=1)
OUTPUT 
   B   C   D
0  1   2   3
1  5   6   7
2  9  10  11



# multi-columns drop
df.drop(['B', 'C'], axis=1)
OUTPUT 
   A   D
0  0   3
1  4   7
2  8  11



# single row drop
df.drop(1, axis=0)
OUTPUT 
   A  B   C   D
0  0  1   2   3
2  8  9  10  11



# multi-row drop
df.drop([1,2], axis=0)
OUTPUT 
   A  B  C  D
0  0  1  2  3




drop_duplicates

import pandas as pd

df = pd.DataFrame({"phone": [909976, 8615246, 2872086, 2273305,2273305,2273305,2273305]})

df
OUTPUT 
     phone
0   909976
1  8615246
2  2872086
3  2273305
4  2273305
5  2273305
6  2273305



df.drop_duplicates('phone',keep='first')
OUTPUT 
     phone
0   909976
1  8615246
2  2872086
3  2273305



df.drop_duplicates('phone',keep='last')
OUTPUT 
     phone
0   909976
1  8615246
2  2872086
6  2273305




dropna

import pandas as pd
import numpy as np

df = pd.DataFrame({'1C':[None, -0.738247,  0.598380,  0.727832],
                   '2C':[None,  0.073079,  1.182290, -0.138224],
                   '3C':[0.554677, -0.530208, -0.397182,  0.990026],
                   '4C':[-0.332384, -1.979684,  0.560655,  0.833487]})
df
OUTPUT 
	1C		2C		3C		4C
0	NaN		NaN		0.554677	-0.332384
1	-0.738247	0.073079	-0.530208	-1.979684
2	0.598380	1.182290	-0.397182	0.560655
3	0.727832	-0.138224	0.990026	0.833487



df.dropna()
OUTPUT 
	1C		2C		3C		4C
1	-0.738247	0.073079	-0.530208	-1.979684
2	0.598380	1.182290	-0.397182	0.560655
3	0.727832	-0.138224	0.990026	0.833487



df.dropna(axis=0)
OUTPUT 
	1C		2C		3C		4C
1	-0.738247	0.073079	-0.530208	-1.979684
2	0.598380	1.182290	-0.397182	0.560655
3	0.727832	-0.138224	0.990026	0.833487



df.dropna(axis=1)
OUTPUT 
	3C		4C
0	0.554677	-0.332384
1	-0.530208	-1.979684
2	-0.397182	0.560655
3	0.990026	0.833487




Analysis

import pandas as pd
df = pd.DataFrame({"Population": [909976, 8615246, 2872086, 2273305,123234,123444,23333,343434],
                   "State": ["Sweden", "United Kingdom", "Italy","Seoul","Suwon", "France","Korea", "Japan"],
                   "Alphabet" : ["a","b","x","d","a","a","b","c"],
                   "rank" : [1,2,3,4,6,5,7,8]})

df
OUTPUT 
   Population           State Alphabet  rank
0      909976          Sweden        a     1
1     8615246  United Kingdom        b     2
2     2872086           Italy        x     3
3     2273305           Seoul        d     4
4      123234           Suwon        a     6
5      123444          France        a     5
6       23333           Korea        b     7
7      343434           Japan        c     8



df.shape
OUTPUT 
(8, 4)



df.columns
OUTPUT 
Index(['Population', 'State', 'Alphabet', 'rank'], dtype='object')



df.index
OUTPUT 
RangeIndex(start=0, stop=8, step=1)



df['Alphabet'].unique()
OUTPUT 
array(['a', 'b', 'x', 'd', 'c'], dtype=object)



df['Alphabet'].value_counts()
OUTPUT 
a    3
b    2
c    1
x    1
d    1
Name: Alphabet, dtype: int64




Statistics

import pandas as pd

df = pd.DataFrame({"Population": [909976, 8615246, 2872086, 2273305,123234,123444,23333,343434],
                   "State": ["Sweden", "United Kingdom", "Italy","Seoul","Suwon", "France","Korea", "Japan"],
                   "Alphabet" : ["a","b","x","d","a","a","b","c"],
                   "rank" : [1,2,3,4,6,5,7,8]})
df
OUTPUT 
	Population	State	          Alphabet	rank
0	909976	  	Sweden	          a	        1
1	8615246	        United Kingdom	  b	        2
2	2872086   	Italy	          x       	3
3	2273305   	Seoul	          d       	4
4	123234	    	Suwon	          a       	6
5	123444	  	France	          a       	5
6	23333	    	Korea	          b       	7
7	343434	        Japan	          c       	8

df.count(axis=0)
OUTPUT 
Population    8
State         8
Alphabet      8
rank          8
dtype: int64

df.count(axis=1)
OUTPUT 
0    4
1    4
2    4
3    4
4    4
5    4
6    4
7    4
dtype: int64

df.corr()
OUTPUT 
	          Population	rank
Population	 1.000000  	-0.573738
rank	      -0.573738    1.000000

df = df.set_index(["Alphabet","rank"]).sort_index()
df
OUTPUT 
               Population           State
Alphabet rank                            
a        1         909976          Sweden
         5         123444          France
         6         123234           Suwon
b        2        8615246  United Kingdom
         7          23333           Korea
c        8         343434           Japan
d        4        2273305           Seoul
x        3        2872086           Italy



df.loc['a'].std()
OUTPUT 
[454165.09584217647]




Several dataframes

Concatenating

import pandas as pd

df1 = pd.DataFrame([[909976, "Sweden1"],
                    [8615246, "United Kingdom1"],
                    [2872086, "Italy1"],
                    [2273305, "France1"]],
                  index=["Stockholm1", "London1", "Rome1", "Paris1"],
                  columns=["Population1", "State1"])

df2 = pd.DataFrame([[909976, "Sweden2"],
                    [8615246, "United Kingdom2"],
                    [2872086, "Italy2"],
                    [2273305, "France2"]],
                  index=["Stockholm2", "London2", "Rome2", "Paris2"],
                  columns=["Population2", "State2"])

df1, df2
OUTPUT 
            Population1           State1
Stockholm1       909976          Sweden1
London1         8615246  United Kingdom1
Rome1           2872086           Italy1
Paris1          2273305          France1 

             Population2           State2
Stockholm2       909976          Sweden2
London2         8615246  United Kingdom2
Rome2           2872086           Italy2
Paris2          2273305          France2


Concat for left and right

df = pd.concat([df1, df2], axis=1)
df
OUTPUT 
            Population1           State1  Population2           State2
London1       8615246.0  United Kingdom1          NaN              NaN
London2             NaN              NaN    8615246.0  United Kingdom2
Paris1        2273305.0          France1          NaN              NaN
Paris2              NaN              NaN    2273305.0          France2
Rome1         2872086.0           Italy1          NaN              NaN
Rome2               NaN              NaN    2872086.0           Italy2
Stockholm1     909976.0          Sweden1          NaN              NaN
Stockholm2          NaN              NaN     909976.0          Sweden2



df = pd.concat([df1, df2], axis=1, ignore_index=True)
df
OUTPUT 
                    0                1          2                3
London1     8615246.0  United Kingdom1        NaN              NaN
London2           NaN              NaN  8615246.0  United Kingdom2
Paris1      2273305.0          France1        NaN              NaN
Paris2            NaN              NaN  2273305.0          France2
Rome1       2872086.0           Italy1        NaN              NaN
Rome2             NaN              NaN  2872086.0           Italy2
Stockholm1   909976.0          Sweden1        NaN              NaN
Stockholm2        NaN              NaN   909976.0          Sweden2



df = pd.concat([df1, df2], axis=1, keys=['C0', 'C1'])
df
OUTPUT 
                    C0                           C1                 
           Population1           State1 Population2           State2
London1      8615246.0  United Kingdom1         NaN              NaN
London2            NaN              NaN   8615246.0  United Kingdom2
Paris1       2273305.0          France1         NaN              NaN
Paris2             NaN              NaN   2273305.0          France2
Rome1        2872086.0           Italy1         NaN              NaN
Rome2              NaN              NaN   2872086.0           Italy2
Stockholm1    909976.0          Sweden1         NaN              NaN
Stockholm2         NaN              NaN    909976.0          Sweden2


Concat for up and down

df = pd.concat([df1, df2], axis=0)
df
OUTPUT 
            Population1  Population2           State1           State2
Stockholm1     909976.0          NaN          Sweden1              NaN
London1       8615246.0          NaN  United Kingdom1              NaN
Rome1         2872086.0          NaN           Italy1              NaN
Paris1        2273305.0          NaN          France1              NaN
Stockholm2          NaN     909976.0              NaN          Sweden2
London2             NaN    8615246.0              NaN  United Kingdom2
Rome2               NaN    2872086.0              NaN           Italy2
Paris2              NaN    2273305.0              NaN          France2



df = pd.concat([df1, df2], axis=0, ignore_index=True)
df
OUTPUT 
   Population1  Population2           State1           State2
0     909976.0          NaN          Sweden1              NaN
1    8615246.0          NaN  United Kingdom1              NaN
2    2872086.0          NaN           Italy1              NaN
3    2273305.0          NaN          France1              NaN
4          NaN     909976.0              NaN          Sweden2
5          NaN    8615246.0              NaN  United Kingdom2
6          NaN    2872086.0              NaN           Italy2
7          NaN    2273305.0              NaN          France2



df = pd.concat([df1, df2], axis=0, keys=['C0', 'C1'])
df
OUTPUT

United Kingdom = UK

               Population1  Population2      State1      State2
C0 Stockholm1     909976.0          NaN     Sweden1         NaN
   London1       8615246.0          NaN         UK1         NaN
   Rome1         2872086.0          NaN      Italy1         NaN
   Paris1        2273305.0          NaN     France1         NaN
C1 Stockholm2          NaN     909976.0         NaN     Sweden2
   London2             NaN    8615246.0         NaN         UK2
   Rome2               NaN    2872086.0         NaN      Italy2
   Paris2              NaN    2273305.0         NaN     France2




Merging




Analysis

import pandas as pd
idx1 = pd.Index([2, 1, 3, 4])
idx2 = pd.Index([3, 4, 5, 6])

idx1.difference(idx2)

Int64Index([1, 2], dtype=’int64’)

idx1.difference(idx2, sort=False)

Int64Index([2, 1], dtype=’int64’)




Input/Output

Input dataset

import pandas as pd

df = pd.read_csv(r'C:\Users\userd\Desktop\dataset\iris.csv')
df.head()
OUTPUT

캡처


import pandas as pd

df = pd.read_csv(r'C:\Users\userd\Desktop\dataset\iris.csv', index_col=0)
df.head()
OUTPUT

캡처


import pandas as pd

df = pd.read_csv(r'C:\Users\userd\Desktop\dataset\iris.csv', index_col='sepal.length')
df.head()
OUTPUT

캡처


import pandas as pd

df = pd.read_csv(r'C:\Users\userd\Desktop\dataset\iris.csv', index_col='variety')
df.head()
OUTPUT

캡처




Output dataset

import pandas as pd

df = pd.DataFrame([[4.78232104, 5.82145535],
                   [6.48127781, 6.33186404],
                   [4.63813463, 5.68560883]])

#saving in excel format
filepath = 'pixel_values.xlsx'
df.to_excel(filepath, index=False)

Covert Data-Type

DataFrame to Series

>>> import pandas as pd

# based on column
>>> df = pd.DataFrame({'phone': [1001, 1002, 1003, 1004, 1005, 1006, 1007]})
>>> type(df)
pandas.core.frame.DataFrame

>>> df = df['phone']
>>> type(df)
pandas.core.series.Series


# based on row
>>> df = pd.DataFrame([[909976, 2872086, 8615246, 2872086]])
>>> type(df)
pandas.core.frame.DataFrame

>>> df = df.loc[0]
>>> type(df)
pandas.core.series.Series




Series to DataFrame

>>> import pandas as pd
>>> s = pd.Series([1,2,3,4,5])
>>> type(s)
pandas.core.series.Series

>>> s= pd.DataFrame(s)
>>> type(s)
pandas.core.frame.DataFrame




DataFrame to numpy

>>> import pandas as pd
>>> df = pd.DataFrame({
        'phone': [1001, 1002, 1003, 1004, 1005, 1006, 1007],
        '이름': ['둘리', '도우너', '또치', '길동', '희동', '마이콜', '영희']})
>>> type(df)
pandas.core.frame.DataFrame

>>> df = df.values
>>> type(df)
numpy.ndarray




List of posts followed by this article

Reference