Matplotlib: user defined plot function print twice

Question

I want to create a plot with a function, it will return a fig so later on I can redisplay it when needed.

The function goes like this:

def simple_plot(ax = None):
    if ax is None:
        fig, ax = plt.subplots()
    a = [1,2,3,4]
    b = [3,4,5,6]
    plt.plot(a, b,'-', color='black')
    return fig

If I run simple_plot(), it would print the plot twice, like this:

Notice: if I run fig = simple_plot(), it will only print once, and I can use fig to reproduce the plot later in Ipython Notebook

How can I make it only print once if I run simple_plot()?
I'm not sure if I defined the function correctly, what would be a good way to define a function to make plot?

Your problem isn't very clear, could you have a read through and try to fix some of the issues? — Jezzamon
– Jezzamon, Commented Feb 16, 2016 at 3:21
@Jezzamon, I have update my question, please see it again, thanks. — ZK Zhao
– ZK Zhao, Commented Feb 16, 2016 at 3:29

Steven C. Howell · Accepted Answer · 2021-02-23 14:20:43Z

23

This is a side effect of the automatic display functionality of the Jupyter Notebooks. Whenever you call plt.plot() it triggers the display of the plot. But also, Jupyter displays the return value of the last line of every cell, so if the figure object is referenced as the last statement of the cell, another display is triggered. If the last statement of the cell is an assignment (fig = simple_plot()), the return value is None and thus a second display is not triggered and you don't get the second plot.

edited Feb 23, 2021 at 14:20

Steven C. Howell

18.9k17 gold badges76 silver badges103 bronze badges

answered Feb 16, 2016 at 3:45

foglerit

8,4369 gold badges52 silver badges70 bronze badges

Sign up to request clarification or add additional context in comments.

2 Comments

cge Over a year ago

This seems very inconvenient for people writing packages that produce custom plots. Is there no way to write a function that returns a plot but doesn't require unusual user behavior to avoid having the plot duplicated?

NuValue Over a year ago

Excellent answer. Just wanted to add that it works fine without adding the semi colon as well. Simply just doing an assignation to object 'fig' in your last line of the cell as @foglerit has stated.

JunPan · Accepted Answer · 2018-03-20 06:03:27Z

14

Just add plt.close() before return, like this:

def simple_plot(ax = None):
    if ax is None:
        fig, ax = plt.subplots()
    a = [1,2,3,4]
    b = [3,4,5,6]
    plt.plot(a, b,'-', color='black')
    plt.close()
    return fig

answered Mar 20, 2018 at 6:03

JunPan

1411 silver badge3 bronze badges

Comments

gsmafra · Accepted Answer · 2016-02-16 03:40:47Z

1

Remove your return statement

Ipython Notebook is plotting once when you execute plt.plot(a, b,'-', color='black') and a second time when you return your fig object to the console.

You could as well keep the return statement, but store the returned value into a variable and plot the figure again just by executing fig.

answered Feb 16, 2016 at 3:40

gsmafra

2,50422 silver badges27 bronze badges

2 Comments

ZK Zhao Over a year ago

This means I can't run simple_plot()? The variable assignment fig = simple_plot() is quite unnecessary when I don't need the fig.

gsmafra Over a year ago

You can but then you don't want a return statement because that will cause the double plot (or you could add the semicolon like jonnat's answer suggested)

Vaibhav Kadam · Accepted Answer · 2022-06-15 17:16:35Z

1

I also encountered this thing. Indeed this is a side effect of the automatic rendering engine embedded in the Jupyter Notebook.

I've found this simple hack. Just use ; after the plot()

for e.g. :

plt.rc("figure",figsize=(8,8))
seasonal_decompose(df,model='additive',period=4).plot();

answered Jun 15, 2022 at 17:16

Vaibhav Kadam

7291 gold badge5 silver badges22 bronze badges

Collectives™ on Stack Overflow

Matplotlib: user defined plot function print twice

4 Answers 4

2 Comments

Comments

2 Comments

Comments

Your Answer

Linked

Hot Network Questions

Collectives™ on Stack Overflow

4 Answers 4

2 Comments

Comments

2 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Linked

Related