Introduction to Pandas

Winter 2025

1

Adrian Salguero

Announcements

• Homework 6 due Friday at 11:59pm
• Coding Practice 8 due tonight at 11:59pm
• Section tomorrow will be a Final Exam Review
• TAs will go over practice problems
• Complete the Course Evaluation if you haven’t already done so
• Friday’s lecture (3/14) will also be a Final Exam Review
• Come with questions!

2

Topics for Today

• What is pandas?
• Basics of pandas
• Installing pandas
• DataFrames
• Series
• Importing and using a csv file in pandas
• Using pandas for visualization

3

Dataset for Today

Kaggle website for dataset: https://www.kaggle.com/datasets/nelgiriyewithana/top-spotify-songs-2023?resource=download

​

CSV file

4

What is pandas?

• Pandas is a Python library (or module) used primarily for data analysis

• Drawbacks
• It can be very inefficient in terms of resource-use, especially when working with incredibly large datasets

• The primary data structure object used in pandas is called the DataFrame and Series
• These objects will give us access to useful methods for working with data

5

Installing pandas

Two common ways to install pandas

​

1. Installing Anaconda
• Pandas comes with Anaconda so no additional installation required
• Write import pandas as pd in your Python file to use

​

• Using Python package installer (pip)
• pip install pandas
• import pandas as pd

6

DataFrames and Series

• A DataFrame is a two-dimensional structure used to store and maintain data
• A Series is a one-dimensional structure that represents a single column or row

7

DataFrame

Series

Reading csv file into DataFrame

• Deciding on how to store csv file data in our program
• Which data structure to use: list, dictionary, set, tuple, etc…
• Parse data correctly (remove whitespace and split on the commas) and store information correctly to use

• Pandas provides us the tool to read csv files (and other files) into a DataFrame
• pandas.read_csv(filename)
• Excel files, Stata files, SQL files

8

Useful pandas commands

• DataFrame.index = returns the index (row) labels
• DataFrame.columns = returns all the column labels
• DataFrame.head(<n = number>) = return the first 5 rows, otherwise if

n = number is defined returns the first number of rows

• DataFrame.tail(n = number) = returns the last 5 rows, otherwise if

n = number is defined, returns the last number of rows

• DataFrame.shape = returns the dimensions of the DataFrame (rows, columns)
• DataFrame.itterows() = allows you to iterate through rows of the DataFrame in tuples → (Index, Series)
• DataFrame.set_index(column) = returns a new DataFrame where index (rows) of panda DataFrame is set to an existing column

9

Slicing and Filtering a DataFrame

• Much like Excel or Google Sheets allows you to filter a spreadsheet, you can do the same in pandas!

​

Check for a single condition

new_df = df[df['column'] == "value"]

​

Check for multiple conditions

new_df = df[(df['column1'] == "value1") & (df['column2']>'value2')]

new_df = df[(df['column1'] == "value1") | (df['column2']>'value2')]

new_df = df[~(df['column2']>'value2')]

​

10

and

or

not

Visualization through pandas

• DataFrames have different plots you can create from them
• DataFrame.plot.bar() = bar graphs
• DataFrame.plot.barh() = horizontal bar graphs
• DataFrame.plot.hist() = histograms
• DataFrame.plot.box() = box plots
• DataFrame.plot.scatter() = scatter plots
• Many others…

​

11