Numpy array element-by-element comparison optimization

Question

Let a be a numpy array of length n. Does the statement a == max(a) calculate the expression max(a) n-times or just one?

You probably want numpy.amax rather than plain Python max. — Gareth Rees
– Gareth Rees, Commented Aug 28, 2014 at 18:59

shx2 · Accepted Answer · 2014-08-28 16:46:57Z

1

It computes max(a) once, then it compares the (scalar) result against each (scalar) element in a, and creates a bool-array for the result.

answered Aug 28, 2014 at 16:46

shx2

64.8k17 gold badges139 silver badges166 bronze badges

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

Comments

Daniel · Accepted Answer · 2014-08-28 21:10:36Z

1

It only evaluates max once. You could test this yourself by writing your own function:

def mymax(x):
    print("Calling mymax.")
    return max(x)

Then try

a == mymax(a)

Daniel

19.6k7 gold badges64 silver badges74 bronze badges

answered Aug 28, 2014 at 16:46

bogatron

19.3k7 gold badges56 silver badges49 bronze badges

Great! Thanks also for the insight about how to test it!