Series  

  • Series is a one-dimensional array-like structure containing homogeneous data.

  • Example of a series: 

      10 	23    56    17    52   61   73    90    26    72

  • Key Points:

    • Homogeneous data (all elements in series has the same data type)

    • Size of a series is immutable

    • Values stored in a series are mutable

  Defining a series object  

  • Syntax: 

      seriesVar = pandas.Series( data, index, dtype, copy)

    The parameters of the constructor are:

    • data:   various forms, e.g.: a list

                   Default:  [ ]

    • index:   index values must be unique and hashable, same length as data.

                   Default:   np.arrange(n), i.e.: [0 .. (n-1)]

    • dtype:   data type.

                   Default:   data type is inferred from data

    • copy:   copy data.

                   Default:   False

  Creating Series objects  

  • How to create an empty series: 

    #import the pandas library and aliasing as pd
import pandas as pd

s1 = pd.Series()
print(s1)                 # Output: Series([], dtype: float64)

  • How to create a series with a Python list: 

    #import the pandas library and aliasing as pd
import pandas as pd

s2 = pd.Series([9, 8])
print(s2)                 

Output:    0    9             (0, 1 are indexes)
           1    8
           dtype: int64

DEMO: progs/series01.py

  Using your own indexes  

  • How to create series with your own indexes: 

    data = ['a', b', 'c']

s1 = pd.Series(data,index=[100,101,99])
print(s1)

Output:
          100    a
          101    b
          99     c
          dtype: object

  • You can use symbols as index: 

    s2 = pd.Series(data, ['x', 'y', 'z'])
print(s2)

Output:
        x    a
	y    b
	z    c
	dtype: object

  Create a Series from dict  

  • A dict can be used as input to construct a series

    • If no index is specified:

      • the dictionary keys are used as index

    • If index is passed:

      • the values in the input data corresponding to the labels in the index will be pulled out

    Example: no index specified 

    data = {'a' : 0.0, 'b' : 1.0, 'c' : 2.0}
s = pd.Series(data)

print(s)

Output:
        a    0.0
	b    1.0
	c    2.0
	dtype: float64

Demo: progs/series03.py

  Create a Series from dict  

  • A dict can be used as input to construct a series

    • If no index is specified:

      • the dictionary keys are used as index

    • If index is passed:

      • the values in the input data corresponding to the labels in the index will be pulled out

    Example: index is passed 

    data = {'a' : 0.0, 'b' : 1.0, 'c' : 2.0}
s = pd.Series(data, ['c', 'b', 'a', 'd'])

print(s)

Output:
        c    2.0
	b    1.0
	a    0.0
	d    NaN

Demo: progs/series03.py

  Accessing Data from Series with Position and with Index  

  • Syntax to access the k-th element in a series by its position k: 

         seriesVar[ k ]

  • Syntax to access an element in a series by its index index: 

         seriesVar[ index ]

    Example: 

    s1 = pd.Series([4,5,6,7,8], index = ['a','b','c','d','e'])

print(s1[0])             ---> 4
print(s1[4])             ---> 8

print(s1['a'])           ---> 4
print(s1['e'])           ---> 8

    Comment:   a series is similar to a Python dict... but can also use positional "index"

  Array slices with positional index  

  • Examples: 

    s = pd.Series([1,2,3,4,5],index = ['a','b','c','d','e'])

print(s[1:3])          
                b    5
		c    6

print(s[1:])
                b    5
		c    6
		d    7
		e    8

print(s[:3])
                a    4
		b    5
		c    6

DEMO: progs/series05.py

  Array slices with a list of indexes  

  • Examples: 

    s = pd.Series([1,2,3,4,5],index = ['a','b','c','d','e'])

print(s[['a','c','d']])

Output:
            a    4
	    c    6
	    d    7

DEMO: progs/series05.py