Pandas›Core· 25 min read

The Pandas Series

A Series is a single labelled column of data — like one column from a spreadsheet.

What you will learn

Create a Series
Use labels (the index) to look up values
See how a Series differs from a list

One labelled column

Pandas (a free Python library for working with tables of data — think of it as a programmable spreadsheet) is the most important tool in data science — it is how you load, clean and explore data. Its two building blocks are the Series (one column) and the DataFrame (a whole table). We start with the Series.

A Series is a list of values plus a label for each value. Those labels are called the index. Think of one column in a spreadsheet, where each cell also has a name.

A Series: values with a labelled index

import pandas as pd

sales = pd.Series([250, 300, 175],
                  index=['Mon', 'Tue', 'Wed'])
print(sales)

Note: Output: Mon 250 Tue 300 Wed 175 dtype: int64 Each number has a label on its left. dtype: int64 means the values are whole numbers.

Look things up by their label

Because each value has a label, you can grab it by name — much friendlier than remembering a position number.

Access by label, and reuse the summary functions

print('Tuesday:', sales['Tue'])    # look up by label
print('average:', sales.mean())    # a Series has the maths functions too
print('best   :', sales.max())

Note: Output: Tuesday: 300 average: 241.66666666666666 best : 300 We asked for Tuesday (300), the average of the three days (about 241.7), and the highest day (300, which was Tuesday). A Series keeps all the array maths.

Keep only the values you want

Just like a NumPy array, you can give a Series a True/False condition in the brackets and it keeps only the rows where the condition is True. This is the exact trick you will use to filter whole tables later.

Filter a Series with a condition

print(sales[sales > 200])      # keep only days that sold over 200

Note: Output: Mon 250 Tue 300 dtype: int64 Wednesday (175) was below 200, so it was dropped. The labels travel with their values, so we still see which days survived.

	Python list	Pandas Series
Access by	Position only (`x[0]`)	Label or position
Has labels?	No	Yes (the index)
Built-in mean/max?	No	Yes
It is really	A simple sequence	A NumPy array + labels

Tip: A Series is basically a NumPy array with labels bolted on — so it keeps all the fast maths (.mean(), .max()) and adds friendly look-up by name.

Q. What is the “index” of a Pandas Series?

Answer: The index is the set of labels (like Mon, Tue, Wed) that lets you look values up by name instead of position.

✍️ Practice

Create a Series of three friends’ ages with their names as the index.
Print one friend’s age by their name, then print the average age.

🏠 Homework

Make a Series of your weekly spending (one value per day, days as the index). Print the total for the week and your biggest spending day.