Changing line color in matplotlib based on logic

Question

I can't seem to figure out how to change linecolor in matplotlib based on some simple logic.

For instance,let's say I have:

import numpy as np
from matplotlib import pyplot as plt

A = [1,2,3,4,5]
B = [2,4,6,8,10]
C = [1,3,5,6,7]
D = [1,2,3,3,3]
combined = [A,B,C,D]

Now, let's say I want matplotlib to plot this as a line graph. Thus, there should be 4 separate lines based on each list in combined.

I want to add the condition if a number in list (of combined) is greater than 5 then the individual line be blue. Else, let the individual line be orange.

How do I go about doing something like this? I know the following would plot it just fine.

np_combined = np.array(combined)
times = np.linspace(0,1,5)
plt.plot(times,np_combined.T)

Would I need a double for loop? I tried more than a few, but seem to end up getting an error each time.

for h in np_combined:
    for k in range(5):
        if k > 5:
            plt.plot(times,k,color = 'blue')
        else:
            plt.plot(times,k,color = 'orange')

Error is EOL while scanning string literal

what kinds did you try and what errors did you get?

rassar
– rassar

2016-12-07 02:27:36 +00:00
Commented Dec 7, 2016 at 2:27 — rassar
– rassar, Commented Dec 7, 2016 at 2:27
edited in my attempt

DudeWah
– DudeWah

2016-12-07 02:34:29 +00:00
Commented Dec 7, 2016 at 2:34 — DudeWah
– DudeWah, Commented Dec 7, 2016 at 2:34

Community · Accepted Answer · 2017-05-23 10:30:12Z

rassar's answer, using a conditional to choose the color (or drawing style) is correct. For simple cases, it's perfectly fine.

For more complex cases, and just to set yourself up for them, there is another option: a decision function. You see these commonly in d3js, Bokeh, and visualization apps.

For a simple case, it's something like:

color_choice = lambda x: 'blue' if x > 5 else 'orange'

for sublist in np_combined:
    plt.plot(times, sublist, color=color_choice(max(sublist)))

Here color_choice could also be a traditional function definition. Using a lambda function is just because it's a short one-liner.

For simple cases, defining a selection function may not be much better than a conditional. But say you also wanted to define a line style, and not use the same conditions as the color choice. E.g.:

for sublist in np_combined:
    largest = max(sublist)
    if largest > 5:
        if largest > 10:
            plt.plot(times, sublist, color='blue', ls='--')
        else:
            plt.plot(times, sublist, color='blue', ls='-')
    else:
        if largest <= 2:
            plt.plot(times, sublist, color='orange', ls='.')
        else:
            plt.plot(times, sublist, color='orange', ls='-')

Now you're in a confusing pickle, because you have so much code for just relatively simple color and line choices. It's repetitive, violating the DRY principle of software engineering, inviting errors.

Decision functions can clean that up greatly:

color_choice = lambda x: 'blue' if x > 5 else 'orange'

def line_choice(x):
    if x > 10: return '--'
    if x > 2:  return '-'
    return '.'

for sublist in np_combined:
    largest = max(sublist)
    plt.plot(times, sublist, 
             color=color_choice(largest)),
             ls=line_choice(largest))

Not only does this clean up the code, localizing the decision logic, it makes it easier to change your color, style, and other choices as your program evolves. The only fly in this ointment is that Python lacks, AFIAK, D3's excellent selection of mapping functions, aka "scales".

This was so informative and really will help me in the future. Thank you so much for your time in writing all that out explicitly.

rassar · Accepted Answer · 2016-12-07 03:01:37Z

1

Based on your attempt, try:

for sublist in np_combined:
    if max(sublist) > 5:
        plt.plot(times,sublist,color = 'blue')
    else:
        plt.plot(times,sublist,color = 'orange')

Also, since your error is that you are missing a end quote (that's what EOL means), the error is probably in another line.

edited Dec 7, 2016 at 3:01

answered Dec 7, 2016 at 2:35

rassar

5,7103 gold badges29 silver badges46 bronze badges

1 Comment

DudeWah Over a year ago

That's great. i'm sad I didn't use max. That solves it. it should be: plt.plot(times,sublist,color = 'b')

Collectives™ on Stack Overflow

Changing line color in matplotlib based on logic

2 Answers 2

1 Comment

1 Comment

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

1 Comment

1 Comment

Your Answer

Sign up or log in

Post as a guest

Related