Latexify

Professional matplotlib plots for LaTeX documents. Transform ugly default plots into publication-ready figures with just two lines of code.

Built on the foundational work by Jack Kelly and enhanced for modern research workflows. This package has been successfully used in multiple publications from our sustainability research lab.

Why Latexify?

Default matplotlib produces plots that look amateur and clash with your beautiful LaTeX documents. Latexify fixes this by providing:

• Font matching - Uses same fonts as your document
• Proper sizing - Automatic figure dimensions for single/double columns
• LaTeX symbols - Perfect math rendering with r'$\sin(x)$'
• Clean axes - Removes chart junk and visual clutter
• No scaling needed - Just use \includegraphics{plot.pdf} directly

Visual Comparison

The difference is dramatic. Here's a complete LaTeX document showing default matplotlib vs. latexify:

Page 1: Before vs. After

Page 2: Double Column Layout

Page 3: Multiple Subplots & Code Examples

Key improvements shown:

• Font matching: Latexify plots use the same fonts as document text
• Clean axes: No chart junk, professional appearance
• LaTeX symbols: Perfect math rendering with $\sin(x)$, $\cos(x)$
• Proper sizing: No manual scaling needed in \includegraphics{}
• Column optimization: Automatic sizing for single and double column layouts

Installation

pip install git+https://github.com/sustainability-lab/latexify.git

Or copy latexify.py to your project.

Quick Start

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

# Setup for your document
latexify.latexify(columns=1)  # or columns=2

# Plot with LaTeX symbols
x = np.linspace(0, 2*np.pi, 100)
plt.plot(x, np.sin(x), label=r'$\sin(x)$')
plt.plot(x, np.cos(x), label=r'$\cos(x)$') 
plt.xlabel(r'$x$ (radians)')
plt.ylabel(r'Amplitude')
plt.legend()

# Clean up and save
latexify.format_axes(plt.gca())
latexify.save_fig('plot.pdf')

In your LaTeX document:

\begin{figure}[htbp]
    \centering
    \includegraphics{plot.pdf}  % No width scaling needed!
    \caption{Beautiful, professional plot.}
\end{figure}

Examples for Different Document Classes

IEEE Conference Format

ACM Proceedings Format

Beamer Presentations

Each example shows how latexify automatically adapts to different academic document formats with appropriate font sizes, spacing, and styling. The examples include:

• Code snippets with proper Python string quotes
• 3 subplot layouts optimized for each format
• Before/after comparisons showing the improvement
• Complete LaTeX source for easy adaptation

API Reference

latexify(columns=1, largeFonts=False, fig_width=None, fig_height=None)

Configure matplotlib for LaTeX output.

Parameters:

• columns: 1 for single column, 2 for double column layouts
• largeFonts: True for presentations, False for papers
• fig_width/fig_height: Override automatic sizing (inches)

format_axes(ax)

Clean up axes by removing chart junk and applying professional styling.

save_fig(filename, **kwargs)

Save figure with optimal settings for LaTeX inclusion.

Document Class Specific Usage

Standard Article

latexify.latexify(columns=1)  # Single column
latexify.latexify(columns=2)  # Double column  

IEEE Format

latexify.latexify(fig_width=3.4, fig_height=2.3)
plt.rcParams.update({'font.size': 8})

ACM Format

latexify.latexify(fig_width=3.3, fig_height=2.2)
plt.rcParams.update({'font.size': 9})

Beamer Presentations

latexify.latexify(largeFonts=True, fig_width=4.5, fig_height=3.0)
# Automatically uses larger fonts suitable for presentations

Finding Column Width for Any Template

Need custom dimensions? Here's how to find your template's column width:

Method 1: LaTeX Commands

Add this to your document and check the output:

\the\columnwidth     % Shows column width in points
\the\textwidth       % Shows text width in points

Convert points to inches: width_inches = points / 72.27

Method 2: Ruler Package

\usepackage{ruler}
\begin{document}
\ruler  % Shows ruler with measurements

Method 3: Common Templates

Template	Single Column	Double Column
Standard Article	3.5"	7.0"
IEEE	3.4"	7.0"
ACM	3.3"	6.75"
Springer	3.35"	6.8"

Then use: latexify.latexify(fig_width=YOUR_WIDTH)

Advanced Tips

Custom Font Sizes

latexify.latexify(columns=1)
plt.rcParams.update({
    'font.size': 9,           # Base font size
    'axes.labelsize': 9,      # Axis labels  
    'legend.fontsize': 8,     # Legend
    'xtick.labelsize': 8,     # Tick labels
    'ytick.labelsize': 8
})

Multiple Plots

fig, axes = plt.subplots(2, 2, figsize=(6.5, 5))
for ax in axes.flat:
    ax.plot(data)
    latexify.format_axes(ax)  # Clean each subplot

Golden Ratio Sizing

Latexify automatically uses the golden ratio (≈1.618) for aesthetically pleasing proportions:

# This gives you golden ratio height automatically
latexify.latexify(fig_width=3.5)  # height = 3.5/1.618 ≈ 2.16"

Requirements

• matplotlib >= 3.0.0
• numpy >= 1.15.0
• LaTeX installation (for math rendering)

License

MIT