Pandas Series | Mastering in Python Pandas Library

pandas.Series

Pandas Series is a One Dimensional indexed array. It is most similar to the NumPy array. pandas.Series is a method to create a series.

Here practically explanation about Series.
For using pandas library in Jupyter Notebook IDE or any Python IDE or IDLE, we need to import Pandas, using the import keyword

import pandas as pd

Here we are using as keyword to short pandas name as “pd

The latest version of Pandas Library is 0.24.2 released on 12 March 2019. To know the version of Jupyter Notebook IDE

pd.__version__
Output >>>  '0.24.2' 

Series is similar to python list but series have additional functionality, methods, and operators, because of these series is advanced than a list.

Methods of Creating a Series

1. Creating series from list

but first, we are creating a list

list_1 = [1, 2, -3, 4.5, 'indian']
print(list_1)
Output >>>   [1, 2, -3, 4.5, 'indian']

Python list stores int, float, string data types

Creating series using the above list

series1 = pd.Series(list_1)
print(series1)
Output >>>
          0         1
          1         2
          2        -3
          3       4.5
          4    indian
          dtype: object

Here it is showing 0 1 2 3 4 is index and 1 2 -3 4.5 Indian are data values.

type(series1) 
Output >>>   pandas.core.series.Series

pandas.core.series.Series means series is a one-dimensional array, which can store indexed data

2. Creating Empty Series

Empty series is like an empty list, we can create empty series using an empty list

empty_s = pd.Series([])
print(empty_s)
Output >>>   Series([], dtype: float64)

3. Creating Series using a different method
List inside the series

series2 = pd.Series([1,2,3,4,5])
print(series2)
Output >>>
          0    1
          1    2
          2    3
          3    4
          4    5
          dtype: int64

in index parameter, default index is start from 0 to n (0,1,2,….n) when index is not identified
Here we are creating series with the index parameter
Index length should have equal to the number of data values, otherwise, it shows error

series2 = pd.Series([1,2,3,4,5], index = ['a', 'b', 'c'])
print(series2)
Output >>>
---------------------------------------------------------------------------
ValueError                                Traceback (most recent call last)
<ipython-input-11-c6475a37a2e3> in <module>
----> 1 series2 = pd.Series([1,2,3,4,5], index = ['a', 'b', 'c'])
      2 print(series2)

C:\ProgramData\Anaconda3\lib\site-packages\pandas\core\series.py in __init__(self, data, index, dtype, name, copy, fastpath)
    247                             'Length of passed values is {val}, '
    248                             'index implies {ind}'
--> 249                             .format(val=len(data), ind=len(index)))
    250                 except TypeError:
    251                     pass

ValueError: Length of passed values is 5, index implies 3

ValueError: Length of passed value is 5, index implies 3
We got an error because we passed 3 indexes for 5 data values

series2 = pd.Series([1,2,3,4,5], index = ['a', 'b', 'c', 'd', 'e'])
print(series2)
Output >>>
          a    1
          b    2
          c    3
          d    4
          e    5
          dtype: int64

We can change index to any numbers, alphabates, names etc.

Above you can see dtype: int64, this means our data type has stored in integer 64 bit.
we can change the data type of series
Changing data type of series (Convert int into a float)

series2 = pd.Series([1,2,3,4,5], index = ['a', 'b', 'c', 'd', 'e'], dtype = float)
print(series2)
Output >>>
          a    1.0
          b    2.0
          c    3.0
          d    4.0
          e    5.0
          dtype: float64

4. Creating series from scalar values
scalar values means single value
e.g. 1, 0.5, ‘indian’

s3_scalar = pd.Series(2)
print(s3_scalar)
Output >>>
          0    2
          dtype: int64

for more data values index should be needed.

s3_scalar = pd.Series(2, index = [1,2,3,4,5])
print(s3_scalar)
Output >>>
          1    2
          2    2
          3    2
          4    2
          5    2
          dtype: int64

5. Creating series from python dictionary

s4_dict = pd.Series({'a':1, 'b':2, 'c':3})
print(s4_dict)
Output >>>
          a    1
          b    2
          c    3
          dtype: int64

Accessing element from series

Pandas Series supports most Python functions.
Now, we are accessing element from series2

print(series2)
Output >>>
          a    1.0
          b    2.0
          c    3.0
          d    4.0
          e    5.0
          dtype: float64

We can access any value or data from series by putting index value

series2[3]
Output >>>
          4.0
series2[4]
Output >>>
          5.0

Slicing series

Here we are slicing series with index value 1 to 4 that means 1 is inclusive(it can be taken) and 4 is exclusive(it can be not taken)

series2[1:4]
Output >>>
          b    2.0
          c    3.0
          d    4.0
          dtype: float64

series can be done by using mathematical operators


Adding two serieses

s5 = pd.Series([1,2,3,4,5])
s6 = pd.Series([1,2,3,4,5])

a = s5 + s6
print(a)
Output >>>
          0     2
          1     4
          2     6
          3     8
          4    10
          dtype: int64

we can also add series using add method

s5.add(s6)
Output >>>
          0     2
          1     4
          2     6
          3     8
          4    10
          dtype: int64

min() operator gives minimum value of particular series

min(a)
Output >>>   2

max() operator gives maximum value

max(a)
Output >>>   10

Conditional operator

If you want to print less than 8 values

a[a < 8]
Output >>>
          0    2
          1    4
          2    6
          dtype: int64

Using drop() function we can eliminate any index value

a.drop(4)
Output >>>
          0    2
          1    4
          2    6
          3    8
          dtype: int64

Now we are printing series6 (s6)

print(s6)
Output >>>
          0    1
          1    2
          2    3
          3    4
          4    5
          dtype: int64
s7 = pd.Series([1,2,3])
print(s7)
Output >>>
          0    1
          1    2
          2    3
          dtype: int64

Pandas have additional functions to fill missing values, it does not show an error when the value is missing. Missing values are shown by NaN.
See below example:


In pandas, we can add the unequal data values series. Here series s6 have 5 data values and s7 have 3 data values, when we perform addition operation it adds successfully

s6 + s7
Output >>>
          0    2.0
          1    4.0
          2    6.0
          3    NaN
          4    NaN
          dtype: float64

Learn more Python Libraries

Python Pandas Tutorial

Python NumPY Tutorial

Python Matplotlib Tutorial

Leave a Reply

Top