Series

Basics

A Series is a straightforward one-dimensional pandas object that is created with indices. There are a lot of 1D arrays, but Series is designed to be compatible with all the pandas functionalities, making things much easier on us.

my_first_series = pd.Series([5, 10, 15, 20], index=['a', 'b', 'c', 'd'])
print(my_first_series['c'])
15

You can also create a Series without passing the index argument. resulting in the default 0-index that is standard for Python at large:

my_second_series = pd.Series([5, 10, 15, 20])
print(my_second_series[2])
15

As we mentioned earlier, other pandas functions can be used in conjunction with Series. In the following example, we use the idxmax function to yield the index of the largest value in the series.

long_list = np.random.randint(low=1, high=10000, size=100)
long_series = pd.Series(long_list)
biggest_value_index = long_series.idxmax()
print(long_series[biggest_value_index])

Knowing how to use a Series is crucial for understanding and manipulating DataFrames properly. While we use the term column to describe the vertical dimension of DataFrames, the data type is a Series; df.A or df['A'] will return a Series, not a list. By extension, we can think of a DataFrame as a collection of 2 or more Series objects.