44 python data frames

A Data frame is a two-dimensional data structure, i.e., data is aligned in a tabular fashion in rows and columns. https://www.tutorialspoint.com/python_pandas/python_pandas_dataframe.htm

In [1]:

import pandas as pd
import numpy as np

In [2]:

from numpy.random import randn
np.random.seed(3)

In [7]:

m1=randn(5,4)

In [8]:

df1=pd.DataFrame(m1,["x","y","z","m","n"],[2008,2019,2018,2019])

In [9]:

df1

Out[9]:

	2008	2019	2018	2019
x	1.788628	0.436510	0.096497	-1.863493
y	-0.277388	-0.354759	-0.082741	-0.627001
z	-0.043818	-0.477218	-1.313865	0.884622
m	0.881318	1.709573	0.050034	-0.404677
n	-0.545360	-1.546477	0.982367	-1.101068

In [16]:

df1=pd.DataFrame(m1,["x","y","z","m","n"],[2008,2019,2018,2020])

In [101]:

df1=pd.DataFrame(m1,' a b c d e '.split(),"r a dsj sk ".split())

In [104]:

df1

Out[104]:

	2018	2019	2018	2019
x	0.985851	-1.472624	-1.874000	0.593433
y	-1.466441	-0.637149	-1.168911	0.284104
z	1.187989	0.357807	0.483570	-0.709170
m	1.388116	-0.455038	0.582384	0.618823
n	1.722225	0.966258	-0.954509	1.881297

In [119]:

df1[[2018]]

Out[119]:

	2018
x	-1.874000
y	-1.168911
z	0.483570
m	0.582384
n	-0.954509

In [10]:

df1[:'x']

Out[10]:

	2008	2019	2018	2019
x	1.788628	0.43651	0.096497	-1.863493

In [17]:

df1

Out[17]:

	2008	2019	2018	2020
x	1.788628	0.436510	0.096497	-1.863493
y	-0.277388	-0.354759	-0.082741	-0.627001
z	-0.043818	-0.477218	-1.313865	0.884622
m	0.881318	1.709573	0.050034	-0.404677
n	-0.545360	-1.546477	0.982367	-1.101068

In [19]:

df1["sum"]=df1[2018]+df1[2019]+df1[2020]+df1[2008]

In [20]:

df1

Out[20]:

	2008	2019	2018	2020	sum
x	1.788628	0.436510	0.096497	-1.863493	0.458143
y	-0.277388	-0.354759	-0.082741	-0.627001	-1.341889
z	-0.043818	-0.477218	-1.313865	0.884622	-0.950279
m	0.881318	1.709573	0.050034	-0.404677	2.236247
n	-0.545360	-1.546477	0.982367	-1.101068	-2.210537

In [22]:

df1.drop("sum",axis=1)

Out[22]:

	2008	2019	2018	2020
x	1.788628	0.436510	0.096497	-1.863493
y	-0.277388	-0.354759	-0.082741	-0.627001
z	-0.043818	-0.477218	-1.313865	0.884622
m	0.881318	1.709573	0.050034	-0.404677
n	-0.545360	-1.546477	0.982367	-1.101068

In [23]:

df1

Out[23]:

	2008	2019	2018	2020	sum
x	1.788628	0.436510	0.096497	-1.863493	0.458143
y	-0.277388	-0.354759	-0.082741	-0.627001	-1.341889
z	-0.043818	-0.477218	-1.313865	0.884622	-0.950279
m	0.881318	1.709573	0.050034	-0.404677	2.236247
n	-0.545360	-1.546477	0.982367	-1.101068	-2.210537

In [26]:

df1.drop("sum",axis=1,inplace=True)

In [27]:

df1

Out[27]:

	2008	2019	2018	2020
x	1.788628	0.436510	0.096497	-1.863493
y	-0.277388	-0.354759	-0.082741	-0.627001
z	-0.043818	-0.477218	-1.313865	0.884622
m	0.881318	1.709573	0.050034	-0.404677
n	-0.545360	-1.546477	0.982367	-1.101068

In [28]:

df1.drop('n')

Out[28]:

	2008	2019	2018	2020
x	1.788628	0.436510	0.096497	-1.863493
y	-0.277388	-0.354759	-0.082741	-0.627001
z	-0.043818	-0.477218	-1.313865	0.884622
m	0.881318	1.709573	0.050034	-0.404677

In [29]:

df1[2008]

Out[29]:

x    1.788628
y   -0.277388
z   -0.043818
m    0.881318
n   -0.545360
Name: 2008, dtype: float64

In [30]:

df1[x]

---------------------------------------------------------------------------
NameError                                 Traceback (most recent call last)
<ipython-input-30-fc78bdb93add> in <module>()
----> 1 df1[x]

NameError: name 'x' is not defined

In [32]:

df1.loc['x']

Out[32]:

2008    1.788628
2019    0.436510
2018    0.096497
2020   -1.863493
Name: x, dtype: float64

In [34]:

df1.iloc[0]

Out[34]:

2008    1.788628
2019    0.436510
2018    0.096497
2020   -1.863493
Name: x, dtype: float64

In [39]:

df1.loc["x",2008]

Out[39]:

1.7886284734303186

In [40]:

df1

Out[40]:

	2008	2019	2018	2020
x	1.788628	0.436510	0.096497	-1.863493
y	-0.277388	-0.354759	-0.082741	-0.627001
z	-0.043818	-0.477218	-1.313865	0.884622
m	0.881318	1.709573	0.050034	-0.404677
n	-0.545360	-1.546477	0.982367	-1.101068

In [41]:

df1.loc[['x','y'],[2018,2019]]

Out[41]:

	2018	2019
x	0.096497	0.436510
y	-0.082741	-0.354759

In [43]:

type(df1.loc[['x','y'],[2018,2019]])

Out[43]:

pandas.core.frame.DataFrame