Section 1.1: Loading Data into Pandas

The initial step in applying Pandas for business intelligence is to import your data. The library offers various functions for this purpose.

Loading Data from Files

Pandas supports multiple file formats, including CSV, Excel, and JSON. To load data from a CSV file, utilize the read_csv() function, as shown below:

import pandas as pd

df = pd.read_csv('data.csv')

Loading Data from Databases

Pandas can also retrieve data from databases using the read_sql() function. For instance, to fetch data from a MySQL database, you can execute:

import pandas as pd

df = pd.read_sql('SELECT * FROM table', con=connection)

Section 1.2: Manipulating Data with Pandas

Once the data is imported, you can manipulate it using various Pandas functions.

Selecting Data

Use the following functions to select data:

• head(): Retrieves the first n rows
• tail(): Retrieves the last n rows
• sample(): Returns a random sample of rows
• loc(): Selects rows by label
• iloc(): Selects rows by position

For example, to fetch the first five rows, you would write:

df.head(5)

To access the last five rows, use:

df.tail(5)

Filtering Data

You can filter your data using methods like:

• query(): Selects rows based on a query
• isin(): Filters rows based on a list of values
• between(): Selects rows within a specified range
• mask(): Filters rows based on a mask
• where(): Selects rows meeting a condition

For instance, to filter rows where the label is 'A', you would use:

df.query('label == "A"')

To filter for specific values, such as 1, 2, or 3, you can use:

df.isin([1, 2, 3])

Sorting Data

To sort your data, employ the sort_values() function. For example, to sort by the value column in ascending order, use:

df.sort_values('value')

To sort in descending order:

df.sort_values('value', ascending=False)

Aggregating Data

Aggregate your data using functions like:

• count(): Counts rows
• mean(): Computes the average value
• median(): Finds the median value
• min(): Identifies the minimum value
• max(): Determines the maximum value

For example, to count the rows in your dataset, use:

df.count()

Grouping Data

Group your data using the groupby() function. For instance, to group by the label column and calculate the mean for each group, you can write:

df.groupby('label').mean()

Section 1.3: Visualizing Data with Pandas

After data manipulation, visualizing the results is essential. Pandas offers multiple functions for creating visual representations of your data.

Plotting Data

You can create various plots using:

• plot(): Generates a line plot
• scatter(): Creates a scatter plot
• bar(): Produces a bar plot
• hist(): Generates a histogram

To create a line plot of your data, use:

df.plot()

For a scatter plot, you can write:

df.plot.scatter()

To save your plots, use the savefig() function, as shown below:

df.plot.savefig('plot.png')

Explore the fundamentals of data analysis with Pandas in this introductory video, which provides a step-by-step guide for beginners.

This updated video tutorial covers comprehensive techniques for utilizing Pandas in data science, ensuring you have the most current information.

In this guide, we explored how to effectively use Pandas for business intelligence, focusing on:

1. Loading data into Pandas
2. Manipulating data with Pandas
3. Visualizing data with Pandas

We hope you found this information valuable.

Before you go:

If you appreciated this guide, please give it a few claps and follow me to receive updates on new publications. Don't hesitate—sign up now to take full advantage of all that Medium has to offer.

About the Author:

Alain Saamego: Software engineer, writer, and content strategist at SelfGrow.co.uk

Email: [email protected]

Follow me on Twitter for more insights and content.