How to put contour labels as legend in a plot

Question

I am plotting contours, using an example from this link (code and output plot below) https://www.geeksforgeeks.org/matplotlib-pyplot-contour-in-python/

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.ticker as ticker
import matplotlib
   
delta = 0.15
x = np.arange(1.5, 2.5, delta)
y = np.arange(1.0, 3.0, delta)
X, Y = np.meshgrid(x, y)
Z = (np.exp(X - Y))
  
  
CS1 = plt.contour(X, Y, Z)
   
fmt = {}
strs = ['1', '2', '3', '4', '5', '6', '7']
for l, s in zip(CS1.levels, strs):
    fmt[l] = s
plt.clabel(CS1, CS1.levels, inline = True,
           fmt = fmt, fontsize = 10)
  
plt.title('matplotlib.pyplot.contour() Example')
plt.show()

The output is:

I want the labels for each contour as a legend in the corner, not along the contour lines. Is there a way?

The legend is created with the following content when referenced from the [reference](https://matplotlib.org/stable/api/contour_api.html). `handles, lables = CS1.legend_elements('y');plt.legend(handles, lables, loc='upper right')` — r-beginners, Jun 29 '22 at 14:25

jylls · Answer 1 · 2022-06-29T14:13:28.220

This post here shows a couple of ways to create a legend for your contours. Specifically, you can use

for i in range(len(cl)): 
  CS1.collections[i].set_label(cl[i].get_text())

to create a custom legend for your contours.

To remove the labels from the contours inside the plot, you can use the solution posted here. It consists in iteratively removing the labels with label.remove()

Below is what the final code looks like:

import numpy as np
import matplotlib.pyplot as plt
import matplotlib.ticker as ticker
import matplotlib
   
delta = 0.15
x = np.arange(1.5, 2.5, delta)
y = np.arange(1.0, 3.0, delta)
X, Y = np.meshgrid(x, y)
Z = (np.exp(X - Y))
  
  
CS1 = plt.contour(X, Y, Z)

fmt = {}

strs = ['1', '2', '3', '4', '5', '6', '7']
for l, s in zip(CS1.levels, strs):
    fmt[l] = s

cl=plt.clabel(CS1, CS1.levels,fmt=fmt, inline = False, fontsize = 10)

plt.title('matplotlib.pyplot.contour() Example')
for i in range(len(cl)): 
  CS1.collections[i].set_label(cl[i].get_text())


for label in cl:
  label.remove()

plt.legend()

And the output gives:

score 1 · Answer 2 · answered Jun 29 '22 at 14:09

Try this:

delta = 0.15
x = np.arange(1.5, 2.5, delta)
y = np.arange(1.0, 3.0, delta)
X, Y = np.meshgrid(x, y)
Z = (np.exp(X - Y))
  
CS1 = plt.contour(X, Y, Z)
   
nm, lbl = CS1.legend_elements()
lbl_ = [i for i in range(1, len(lbl))]
plt.legend(nm, lbl_) 

plt.title('matplotlib.pyplot.contour() Example')
plt.show()

How to put contour labels as legend in a plot

2 Answers2