pandas-tutorial

Tutorial on using Pandas, a popular data analysis framework in Python, to perform data analysis tasks on structured data.

Contents

• Repository Structure
• Dependencies
• Disclaimer

Repository Structure

This repository is structured as follow:

contents/   : contains Python scripts for instructor's demonstration purposes during training session
exercises/  : contains Python scripts with instructions for participants to practice

Class contents are structured in such a way that core functionalities of pandas are exposed to participants in a step-by-step and functional manner. Following contains short description of subject covered for each scripts:

• 01_intro.ipynb: Introduces Series and DataFrame, two core data structures of pandas
• 02_load_and_save.ipynb: Importing and exporting files/datasets
• 03_data_manipulation.ipynb: Indexing and slicing data, loc and iloc methods
• 04_EDA.ipynb: General methods to perform exploratory data analysis
• 05_data_cleaning.ipynb: Performing missing data identification and imputation
• 06_data_analysis.ipynb: Performing more in-depth data analysis using rolling window, grouping method and more
• 07_data_visualization.ipynb: Depicting data in various visualizations

Dependencies

Isolating each environment for different projects is the best practice. One of the way you can create virtual environment is by using Python's native virtualenv module. At this directory's root, execute the following to create a virtual environment:

python3 -m venv venv

Commands to activate virtual environment varies according to your OS. Use the following for Linux:

source venv/bin/activate

Use the following for Windows:

venv\Scripts\activate

A (venv) appearing in front of your system path indicates that the virtual environment is successfully activated, as shown below:

(venv) C:\Users\User\pandas-tutorial>

Finally, execute the following to install the dependencies for this lab into your activated virtual environment:

pip install -r requirements.txt

Disclaimer

All the dataset used in this repository does not belong to the authors but rather are open-sourced datasets found online. Attached are the URLs for each and every dataset:

• ct.json dataset source
• airlines.csv dataset source
• Pokemon.csv dataset source
• Titanic.csv dataset source