mpl-lego

mpl-lego is a small collection of reusable utilities for building and styling Matplotlib figures. It includes helpers for subplot labels, covariance ellipses, colorbars, legends, scatter plots, colors, and consistent axis styling.

Installation

Install the latest release from PyPI:

python -m pip install mpl-lego

mpl-lego supports Python 3.11 and newer and depends on Matplotlib and NumPy.

Quick start

import matplotlib.pyplot as plt
import numpy as np

from mpl_lego.axes import style_panel_axis
from mpl_lego.labels import apply_subplot_labels
from mpl_lego.scatter import tighten_scatter_plot

rng = np.random.default_rng(42)
observed = rng.normal(size=100)
predicted = observed + rng.normal(scale=0.35, size=100)

fig, ax = plt.subplots(figsize=(5, 5))
ax.scatter(observed, predicted, alpha=0.7)

style_panel_axis(ax, grid_axis="both")
tighten_scatter_plot(ax, color="black", linestyle="--")
apply_subplot_labels(ax, labels=["a"])

ax.set(xlabel="Observed", ylabel="Predicted")
fig.tight_layout()
plt.show()

Utilities

All public functions include NumPy-style docstrings with their parameters and return values.

Development

Development uses uv for dependency management, virtual environments, testing, and package builds:

git clone https://github.com/pssachdeva/mpl-lego.git
cd mpl-lego
uv sync
uv run pytest
uv build

uv sync creates the project environment from uv.lock and installs the development dependency group automatically.

Three known regressions are tracked in the issue tracker. Their tests are marked as expected failures until the corresponding fixes land.

License

mpl-lego is distributed under the MIT License.