How to add different graphs (as an inset) in another python graph


I’d like to make a graph like that:

Ilustrative Figure from google

the problem is, I’ve got the data from some external files, and I can make the background graph, but I have no idea how to add another graph inside of the one that I already have and change the data to have different results in both of them:

Below I am adding the code I am using to do the background graph.
Hope someone can help.

from __future__ import division
import numpy as np
import matplotlib.pyplot as plt

font = {'family':'serif','size':16}


plt.plot(data[:,0],data[:,6],linewidth = 3,label='B$_0$ = 1.5 T d',linestyle= '--', color='black')
Asked By: Mac



There’s more than one way do to this, depending on the relationship that you want the inset to have.

If you just want to inset a graph that has no set relationship with the bigger graph, just do something like:

import matplotlib.pyplot as plt

fig, ax1 = plt.subplots()

# These are in unitless percentages of the figure size. (0,0 is bottom left)
left, bottom, width, height = [0.25, 0.6, 0.2, 0.2]
ax2 = fig.add_axes([left, bottom, width, height])

ax1.plot(range(10), color='red')
ax2.plot(range(6)[::-1], color='green')

enter image description here

If you want to have some sort of relationship between the two, have a look at some of the examples here:

This is useful if you want the inset to be a “zoomed in” version, (say, at exactly twice the scale of the original) that will automatically update as you pan/zoom interactively.

For simple insets, though, just create a new axes as I showed in the example above.

Answered By: Joe Kington

You can do this with inset_axes method (see docs):

from mpl_toolkits.axes_grid1.inset_locator import inset_axes
inset_ax = inset_axes(parent_axes,
                    width="30%", # width = 30% of parent_bbox
                    height=1., # height : 1 inch

See this example for a full demo.

Answered By: pms
Categories: questions Tags: , ,
Answers are sorted by their score. The answer accepted by the question owner as the best is marked with
at the top-right corner.