Python equivalent to 'hold on' in Matlab

Question

Is there an explicit equivalent command in Python's matplotlib for Matlab's hold on? I'm trying to plot all my graphs on the same axes. Some graphs are generated inside a for loop, and these are plotted separately from su and sl:

import numpy as np
import matplotlib.pyplot as plt

for i in np.arange(1,5):
    z = 68 + 4 * np.random.randn(50)
    zm = np.cumsum(z) / range(1,len(z)+1)
    plt.plot(zm)
    plt.axis([0,50,60,80])

plt.show()

n = np.arange(1,51)
su = 68 + 4 / np.sqrt(n)
sl = 68 - 4 / np.sqrt(n)

plt.plot(n,su,n,sl)

plt.axis([0,50,60,80])
plt.show()

Possible dupe of [multiple-plot-in-one-figure-in-python](http://stackoverflow.com/questions/21254472/multiple-plot-in-one-figure-in-python/21254861) — Kraay89, Jan 30 '14 at 19:46
@Cody Piersall - It might work for others, but apparently not for my example. — Medulla Oblongata, Jan 30 '14 at 19:48
I hope this helps since this worked for me: [LINK](https://stackoverflow.com/questions/12670101/matplotlib-ion-function-fails-to-be-interactive) — troymyname00, Feb 07 '17 at 03:01
`plt.hold(True)` is deprecated in latest version of `matplotlib`. Use plt.show() in the program — , Sep 09 '20 at 09:30

score 62 · Accepted Answer · answered Jan 30 '14 at 19:49

62

Just call plt.show() at the end:

import numpy as np
import matplotlib.pyplot as plt

plt.axis([0,50,60,80])
for i in np.arange(1,5):
    z = 68 + 4 * np.random.randn(50)
    zm = np.cumsum(z) / range(1,len(z)+1)
    plt.plot(zm)    

n = np.arange(1,51)
su = 68 + 4 / np.sqrt(n)
sl = 68 - 4 / np.sqrt(n)

plt.plot(n,su,n,sl)

plt.show()

answered Jan 30 '14 at 19:49

Alvaro Fuentes

16,937
4
56
68

8

Note: This doesn't work in the case of interactive input from the command line. Works fine if run within a script. – ZZZ Jul 21 '17 at 20:25

mapsa · Answer 2 · 2016-04-19T09:34:35.197

23

You can use the following:

plt.hold(True)

edited Apr 19 '16 at 09:34

answered Jun 20 '14 at 02:32

mapsa

464
4
5

20

This function is deprecated, Any alternative solution like this? – Jyotirmay Jun 29 '17 at 14:22
3

@Jyotirmay The hold function in matplotlib does not serve any purpose and was hence deprecated. pyplot always acts as if `hold` was set to True anyways. – ImportanceOfBeingErnest Apr 17 '19 at 15:51

freude · Answer 3 · 2018-01-06T12:55:22.020

19

The hold on feature is switched on by default in matplotlib.pyplot. So each time you evoke plt.plot() before plt.show() a drawing is added to the plot. Launching plt.plot() after the function plt.show() leads to redrawing the whole picture.

edited Jan 06 '18 at 12:55

answered Jul 21 '17 at 10:52

freude

3,632
3
32
51

1

How would you get around this when using IPython? – ThatNewGuy Apr 06 '21 at 12:11
1

use backend %matplotlib inline and all plt.plot() within one cell will be output in the same frame. If you put plt.show() between plt.plot() functions within a cell it will create more frames, new one for each plt.show() – freude Apr 07 '21 at 04:45

score 0 · Answer 4 · answered Jan 26 '18 at 08:39

check pyplot docs. For completeness,

import numpy as np
import matplotlib.pyplot as plt

#evenly sampled time at 200ms intervals
t = np.arange(0., 5., 0.2)

# red dashes, blue squares and green triangles
plt.plot(t, t, 'r--', t, t**2, 'bs', t, t**3, 'g^')
plt.show()

score 0 · Answer 5 · answered Oct 21 '21 at 11:55

Use plt.sca(ax) to set the current axes, where ax is the Axes object you'd like to become active.

For example:

In a first function: import numpy as np import matplotlib.pyplot as plt

for i in np.arange(1,5):
    z = 68 + 4 * np.random.randn(50)
    zm = np.cumsum(z) / range(1,len(z)+1)
    plt.plot(zm)
    plt.axis([0,50,60,80])

plt.show()

In the next function: def function2(...., ax=None)

if ax is None:
    fig, ax = plt.subplots(1)
else:
    plt.sca(ax)

n = np.arange(1,51)
su = 68 + 4 / np.sqrt(n)
sl = 68 - 4 / np.sqrt(n)

plt.plot(n,su,n,sl)

plt.axis([0,50,60,80])
plt.show()

Python equivalent to 'hold on' in Matlab

5 Answers5

Linked

Related