Installation and Import¶

Matplotlib is the foundational plotting library for Python scientific computing.

Mental Model

Matplotlib has one required install (matplotlib) and one standard import convention (import matplotlib.pyplot as plt). The plt alias is universal -- virtually every tutorial, Stack Overflow answer, and textbook uses it. Memorize this single import line and you are ready to plot.

Matplotlib Setup Layers

Matplotlib operates in layers, each controlling a different aspect:

text 1. Environment (installation, backend) → where plots render 2. Defaults (style sheets, rcParams) → how plots look 3. Structure (subplots, GridSpec) → how plots are arranged 4. Content (plot, scatter, etc.) → what data is shown

This section covers layers 1--2. The backend is the rendering engine that converts plotting instructions into pixels or files — Matplotlib separates plotting logic (your Python code) from rendering (the backend), which is why the same code works in Jupyter, scripts, and terminals.

Installation¶

Install via pip:

bash pip install matplotlib

Or via conda:

bash conda install matplotlib

Import Convention¶

The standard import convention uses plt as the alias:

python import matplotlib.pyplot as plt

For numerical data, NumPy is typically imported alongside:

python import matplotlib.pyplot as plt import numpy as np

Version Check¶

python import matplotlib print(matplotlib.__version__)

Backend Configuration¶

Matplotlib uses backends to turn your plotting commands into actual pixels. The rendering pipeline is:

text Python code → pyplot API → Backend → Renderer → Output (screen or file)

The backend determines where and how plots appear. Common backends include:

python import matplotlib matplotlib.use('TkAgg') # Must be called before importing pyplot

In Jupyter notebooks:

python %matplotlib inline # Static images embedded in notebook %matplotlib notebook # Interactive (zoom, pan)

Common Gotchas

Plot window doesn't appear → wrong backend for your environment. Try matplotlib.use('TkAgg') or matplotlib.use('Agg') for headless servers.
Jupyter shows nothing → add %matplotlib inline at the top of the notebook.
Script vs Jupyter behavior differs → scripts need plt.show() to display; Jupyter renders automatically at the end of each cell.

Verifying Installation¶

```python import matplotlib.pyplot as plt import numpy as np

x = np.linspace(0, 2 * np.pi, 100) y = np.sin(x)

plt.plot(x, y) plt.show() ```

If a plot window appears (or renders inline in Jupyter), the installation is successful.

Key Takeaways¶

Install with pip install matplotlib or conda install matplotlib
Import as import matplotlib.pyplot as plt
Configure backend before importing pyplot if needed

Exercises¶

Exercise 1. Write the standard import statement for Matplotlib's pyplot module and NumPy. Then write code that checks and prints the Matplotlib version.

Solution to Exercise 1

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

print(f"Matplotlib version: {matplotlib.__version__}")

Exercise 2. Explain the difference between pip install matplotlib and conda install matplotlib. When would you use each?

Solution to Exercise 2

pip installs from PyPI and works in any Python environment. conda installs from the Anaconda repository and manages non-Python dependencies (C libraries, system packages) automatically. Use pip in plain virtual environments (venv). Use conda when working inside Anaconda/Miniconda environments, especially if you also need compiled scientific libraries like MKL-optimized NumPy.

Exercise 3. Write code that creates a simple test plot to verify Matplotlib is installed correctly. Plot \(y = \sin(x)\) for \(x \in [0, 2\pi]\).

Solution to Exercise 3

import matplotlib.pyplot as plt
import numpy as np

x = np.linspace(0, 2 * np.pi, 100)
y = np.sin(x)

fig, ax = plt.subplots()
ax.plot(x, y)
ax.set_title('Verification: y = sin(x)')
ax.set_xlabel('x')
ax.set_ylabel('sin(x)')
plt.show()

Exercise 4. List three Matplotlib backends and explain when you would use each. Write one line of code showing how to set the backend.

Solution to Exercise 4

Three common backends:

TkAgg — default interactive backend on most systems; opens a Tk window for pan/zoom.
Agg — non-interactive, renders to PNG files; use on headless servers or in scripts that only save figures.
nbAgg / inline — for Jupyter notebooks; %matplotlib inline embeds static PNGs in cells.

Setting the backend in code (must be called before import matplotlib.pyplot):

import matplotlib
matplotlib.use('Agg')  # headless rendering