My personal python matplotlib wrapper for creating publication-level scientific figures with minimal effort.
-
Dependence: A recent version of numpy, matplotlib, and LaTeX.
-
Install: Include src/ folder as the $PYTHONPATH environment in .bashrc; or simply copy src/columnplots.py to the same folder as the plotting script.
-
Example in examples/simple.py illustrates how to use this wrapper to reduce the code effort for generating the publication-level figures for scientific journals.
import numpy as np
# our wrapper
import columnplots as clp
# turn this parameter to false if LaTeX is not installed
if_latex=False
# prepare x data for plotting
x = np.linspace(0, 10, 10)
# prepare y data for plotting
y = (x**2 - x + 3)*1e4
# prepare input values for scientific plotting
xs = [x]*2
ys = [y]*2
colors = ["ko", "r--"]
labels = ["$y_1$", "$y_2$"]
# do plotting
# 1. Initialize a 2x2 figure
axes = clp.initialize(col=2, row=2,
width=6.4, height=6.4*0.618,
LaTeX=if_latex,
labelthem=True, labelsize=11, labelthemPosition=[0.1, 0.95],
fontsize=10)
# 2. Plot each figure
clp.plotone(xs, ys, axes[0,0], labels=labels, colors=colors, lw=1,
ylabel="$y$ value", showlegend=True, legendloc="lower right",
yscientificAtLabel=True)
clp.plotone(xs, ys, axes[0,1], labels=labels, colors=colors, lw=1, showlegend=False,
yscientificAtLabel=True)
clp.plotone(xs, ys, axes[1,0], labels=labels, colors=colors, lw=1, showlegend=False,
ylabel="$y$ value", xlabel="$x$ value",
yscientificAtLabel=True)
clp.plotone(xs, ys, axes[1,1], labels=labels, colors=colors, lw=1, showlegend=False,
xlabel="$x$ value",
yscientificAtLabel=True)
# 3. tune the spacing between subplots and save file as example.png
clp.adjust(tight_layout=True, savefile="example.png")After running this python script, we obtain the following figure:
If we set if_latex=True, the figure would be:
This wrapper has been used in several other Github projects, including