Jupyter Notebook Interface

What is Jupyter Notebook?

Jupyter Notebook is an interactive environment where you can mix:

• Code (Python)
• Text/notes (Markdown)
• Outputs (tables, charts, images)

It’s widely used for:

• Exploratory data analysis (EDA)
• Learning and experimentation
• Quick visualization
• Sharing analysis as a single document

Launching Jupyter Notebook

With Anaconda

1. Open Anaconda Navigator
2. Choose your environment (optional)
3. Click Launch on Jupyter Notebook

From terminal

Activate your environment first, then run:

jupyter notebook

A browser tab opens with the Notebook dashboard.

The Notebook dashboard (files home)

The first screen usually shows:

• A list of files/folders
• Buttons to create new notebooks

Recommended structure for your data analytics projects:

• A folder per project (e.g., titanic-eda/)
• A data/ folder for CSV files
• A notebooks/ folder for .ipynb notebooks
• A reports/ folder for exported HTML/PDF (optional)

Creating a new notebook

From the dashboard:

• Click New → Python 3 (ipykernel) (or similar)

A new .ipynb opens in a new tab.

Understanding cells

Cells are the core concept of notebooks.

Code cell

• Runs Python code
• Produces outputs

Example:

import math
math.sqrt(81)

Markdown cell

• Used for notes and documentation
• Supports headings, lists, links, tables, etc.

Example (Markdown cell):

# Titanic EDA
This notebook explores passenger survival factors.

Running cells

• Run current cell: Shift + Enter
• Run and insert new cell below: Alt + Enter
• Run without moving: Ctrl + Enter

When a cell runs:

• In [1]: indicates the execution order
• Outputs appear below the cell

Kernels (the “engine” behind the notebook)

A kernel is the process that runs your code.

Key idea:

• If you restart the kernel, you lose variables in memory.
• You must run cells again to recreate them.

Common kernel actions:

• Restart: clears memory and restarts Python
• Interrupt: stops long-running code

In the menu:

• Kernel → Interrupt
• Kernel → Restart

Saving notebooks

A notebook is stored as a .ipynb file (JSON-based format).

• Save: Ctrl + S
• Rename: click the title at the top

Visualize it

Each time you run a cell, the kernel executes it and hands the result back to the notebook document.

  flowchart LR
  A["Write code in a cell"] --> B["Press Shift+Enter"]
  B --> C["Kernel executes code"]
  C --> D["Kernel returns result"]
  D --> E["Output shown below the cell"]
  D --> F["In [n] execution count updates"]

A recommended data analytics notebook workflow

1. Import libraries (NumPy, Pandas, Matplotlib)
2. Load data (CSV/Excel/SQL)
3. Inspect data (head(), info(), describe())
4. Clean data (missing values, duplicates, types)
5. Analyze (groupby, aggregates)
6. Visualize (plots)
7. Summarize findings (Markdown cells)

Helpful settings

Auto-reload imports (optional)

In a notebook, you can auto-reload local modules:

%load_ext autoreload
%autoreload 2

Check your understanding

Next

Continue to: Jupyter Shortcuts & Magic Commands to speed up your workflow and learn % / %% notebook magics.

🧪 Try It Yourself

Exercise 1 – Track the Execution Counter

Press Run to see what your code prints.

Exercise 2 – Classify a Cell as Code or Markdown

Press Run to see what your code prints.

Exercise 3 – Restarting the Kernel Clears Memory

Press Run to see what your code prints.