Matplotlib - Formatting Axes

Matplotlib - Formatting Axes Appearance

Beyond setting limits and ticks, Matplotlib offers control over various aesthetic aspects of the axes, such as scale (linear vs. logarithmic), label padding, and the appearance of the axis lines themselves (spines).

Logarithmic Scale

When dealing with data that spans several orders of magnitude, a standard linear scale can compress variations in smaller values, making them hard to see. A logarithmic scale is often more appropriate in such cases.

You can set the scale of the x-axis or y-axis (or both) to logarithmic using ax.set_xscale('log') or ax.set_yscale('log').

Label Padding

Sometimes, the default spacing between an axis label (e.g., ‘X Axis Title’) and the tick labels might not be sufficient, or you might want to adjust it for aesthetic reasons. You can control this spacing using the labelpad parameter within the ax.set_xlabel() or ax.set_ylabel() functions, or by accessing the axis object directly.

Axis Spines

Spines are the lines that form the border of the plotting area, connecting the tick marks. By default, Matplotlib draws spines on all four sides (top, bottom, left, right). You can customize their visibility, color, and line width, or even move their position.

Spines are accessed via the ax.spines dictionary (e.g., ax.spines['top'], ax.spines['bottom']).

Example demonstrating these formatting options:

import matplotlib.pyplot as plt
import numpy as np

# Prepare data
x = np.linspace(1, 10, 100)
y_exp = np.exp(x)
y_pow = x**2

fig, axes = plt.subplots(1, 2, figsize=(12, 5))

# --- Plot 1: Linear Scale, Label Padding, Spine Formatting ---
ax1 = axes[0]
ax1.plot(x, y_exp, label=r'$e^x$')
ax1.plot(x, y_pow, label=r'$x^2$')
ax1.set_title("Linear Scale with Spine Formatting")

# Adjust label padding
ax1.set_xlabel("X Axis", fontsize=12, labelpad=15) # Increased padding
ax1.set_ylabel("Y Axis", fontsize=12, labelpad=10)

# Customize spines
ax1.spines['top'].set_visible(False)    # Hide top spine
ax1.spines['right'].set_color('red')    # Make right spine red
ax1.spines['right'].set_linewidth(2)    # Make right spine thicker
ax1.spines['bottom'].set_linestyle('--') # Make bottom spine dashed

ax1.legend()
ax1.grid(True, alpha=0.5)

# --- Plot 2: Logarithmic Scale (Y-axis) ---
ax2 = axes[1]
ax2.plot(x, y_exp, label=r'$e^x$')
ax2.plot(x, y_pow, label=r'$x^2$')
ax2.set_title("Logarithmic Scale (Y-axis)")
ax2.set_xlabel("X Axis", fontsize=12)
ax2.set_ylabel("Y Axis (log scale)", fontsize=12)

# Set y-axis scale to logarithmic
ax2.set_yscale('log')

# Default spines, but add grid (useful with log scale)
ax2.grid(True, which='both', linestyle='--', alpha=0.5) # Grid for major and minor ticks
ax2.legend()

plt.tight_layout()
plt.show()

In the first plot, we hide the top spine, color the right spine red and make it thicker, and make the bottom spine dashed. We also increase the padding for the x-axis label. In the second plot, we set the y-axis to a logarithmic scale using ax2.set_yscale('log'), which is often helpful when plotting exponential data alongside polynomial data, as it makes both trends more visible across their ranges.