Working Efficiently with Jupyter Notebooks

Code Cells & Markdown Cells

Jupyter notebooks consist of two main cell types: Code cells (executable Python) and Markdown cells (text, equations, explanations).

The key idea behind notebooks is literate programming: mixing narrative text, equations and executable code in a single document. This makes them ideal for exploration, teaching and reporting, but you still need clean structure so that the notebook can be run from top to bottom without hidden state.

# Example code cell
import numpy as np

data = np.random.randn(5)
print("Mean:", data.mean())

# Markdown cell (in the notebook, not in Python)
# ## Section title
# - Bullet 1
# - Bullet 2
# Equation: \\( y = wx + b \\)

Useful Magic Commands

Jupyter provides magic commands (starting with % or %%) to make your workflow faster.

# Time how long a cell takes to run
%timeit sum(range(1_000_000))

# Show plots inline
%matplotlib inline

# Run a shell command
!ls -lh

# Load an extension (e.g., autoreload)
%load_ext autoreload
%autoreload 2

Best Practices for Clean Notebooks

• Keep notebooks linear: re‑run all cells from top to bottom.
• Avoid very long notebooks; split into sections or multiple notebooks.
• Use Markdown headings and text to explain your steps and results.
• Move reusable functions into .py modules and import them.
• Commit notebooks to Git with cleared outputs when possible.