Python and Jupyter Notebook

Introduction to Python

Python is a powerful general-purpose programming language widely used in data science and machine learning.
Python was created in 1991 by Guido van Rossum at Centrum Wiskunde & Informatica (CWI) in the Netherlands.
Python emphasizes code readability and simplicity with a clean, consistent syntax.
Python is free and open source, with a massive ecosystem of libraries and an extensive global community.
The language is stewarded by the Python Software Foundation (PSF) and maintained by a large open-source contributor base.
PyPI (the Python Package Index) hosts community-contributed packages you can install to extend Python.

Why learn Python for Data Science

Python is open source and freely available for Windows, macOS, and Linux.
Rich data-science stack:
- NumPy for fast numerical computing
- pandas for data manipulation (similar to R’s data frames)
- Matplotlib and Seaborn for visualization (and Plotly for interactivity)
- scikit-learn for machine learning
One of the most popular languages for modern machine learning and AI applications in both academia and industry.

Python and the “two-language problem”

Historically, teams prototype in a high-level language (e.g., R or SAS) and then re-implement in a systems language (e.g., C++/Java) for production.
Python increasingly covers both roles: interactive research/prototyping and production services.
When you need extra performance, Python can call into optimized code via Numba, Cython, or C/C++ extensions-so you don’t have to switch languages.

Download & Install Python

Official Python downloads: https://www.python.org/downloads/
For data science, we recommend the Anaconda distribution, which includes most essential packages and tools:
- Download: https://www.anaconda.com/download
To install:
1. Run the downloaded installer.
2. Follow the on-screen instructions.
In this course, we’ll primarily use Anaconda to run Python locally.
After installing, verify Python with python --version (or python3 --version on macOS/Linux; py -3 --version on Windows). Many systems use python3 by default, while Windows commonly uses py -3.

Python Interface (Python REPL)

If you type python (or python3) in your terminal/command prompt, you’ll get an interactive prompt (>>>) that can execute Python statements immediately.
This interactive interpreter is often called the Python REPL (also “Python console/shell”).

REPL stands for Read–Eval–Print Loop:
1. Read the code you type
2. Evaluate it when you press Enter
3. Print the result
4. Loop back to step 1

Scripts vs. Notebooks

A Python script is a plain-text file (typically .py) containing Python code.
Popular editors/IDEs for scripts include VS Code, PyCharm, Spyder, and Thonny.
A Jupyter notebook (.ipynb) is an interactive document that mixes code, text, math, and outputs.
JupyterLab is the modern interface for creating and working with notebooks (and other files) in your browser.

Launch JupyterLab

From Anaconda Navigator: click JupyterLab.
From a terminal/command prompt:
- jupyter lab (recommended)
- jupyter notebook (classic interface)
This opens JupyterLab in your default web browser. To stop it, return to the terminal where it’s running and press Ctrl+C.

In this course, we will use JupyterLab for an enhanced notebook experience.

Working with JupyterLab

Within the JupyterLab window, you can launch a:

Notebook
Console
Terminal
Text Editor, and more

Creating and Opening Notebooks

To create a notebook, click the Notebook icon in the JupyterLab Launcher.
JupyterLab creates a new notebook (e.g., Untitled.ipynb) using the IPython kernel (Python).
To open an existing notebook, locate it in the Files pane, then double-click (or right-click → Open).

Notebook Building Blocks

A Jupyter notebook consists of cells. The two main types are:
1. Code cells — for Python code
2. Markdown cells — for narrative text and math
Use Shift+Enter to run the current cell (then move to the next).
In Markdown cells you can write:
- Regular text (headings, lists, etc.)
- LaTeX math, e.g. $\\alpha+\\beta$ for inline or $$...$$ for display equations
- Basic HTML if needed

Working Directory in Jupyter

JupyterLab typically starts in the directory you launched it from.
New notebooks inherit the folder in which they are created/saved.
Check the current working directory in Python:

import os
os.getcwd()

Change the working directory in Python:

import os
os.chdir("path/to/directory")  # e.g., "C:/Users/you/projects" on Windows or "/Users/you/projects" on macOS/Linux

IPython/Jupyter also provides magics (available in notebooks and IPython consoles):

%pwd                 # show current working directory
%cd /path/to/folder  # change directory
%ls                  # list files in current directory

(Magics are not standard Python; they only work in IPython/Jupyter.)

Listing Files in the Current Directory

Using Python:

import os
os.listdir()   # returns a list of file/folder names

Using Jupyter magics (IPython):

%ls

Python Packages for Data Science

Core packages you’ll use frequently:
- NumPy — numerical arrays, linear algebra, fast vectorized operations
- pandas — data frames, grouping, reshaping, joins, time series
- Matplotlib / Seaborn — static visualization
- Plotly — interactive visualization
- scikit-learn — machine learning models and utilities
Installing packages inside a notebook (preferred magic):

%pip install package-name

Installing packages in a terminal:

# pip
pip install package-name
# or explicitly pip3 if needed
pip3 install package-name

# conda (if using Anaconda/Miniconda):
conda install package-name

Using packages after installation (each new session needs imports):

import pandas as pd
import numpy as np
import matplotlib.pyplot as plt

A quick, fun test: emoji

Try installing the emoji package and using it:

%pip install emoji

import emoji
emoji.emojize("Python is :thumbs_up:", language="alias")
# Alternative common name:
emoji.emojize("Python is :thumbsup:")