Conditional summation in python

Question

I have a numpy 2d array (8000x7200). I want to count the number of cells having a value greater than a specified threshold. I tried to do this using a double loop, but it takes a lot of time. Is there a way to perform this calculation quickly?

piRSquared · Accepted Answer · 2017-06-08 15:15:47Z

4

Assume your variables are defined as

np.random.seed([3,1415])
a = np.random.rand(8000, 7200)
threshold = .5

Then use sum
*(a > threshold) is a boolean array indicating every instance of a cell being greater than some threshold. Since boolean values are a sub-class of int, with False as zero and True as one, we can easily sum them up. numpys sum sums over the entire array by default.

(a > threshold).sum()
28798689

edited Jun 8, 2017 at 15:15

answered Jun 8, 2017 at 15:08

piRSquared

296k68 gold badges509 silver badges654 bronze badges

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

4 Comments

Ma0 Over a year ago

this is almost spot on 50%

piRSquared Over a year ago

@Ev.Kounis that's by nature of the example I chose.

Ma0 Over a year ago

I know but I am still amused by it xD

Meryem Over a year ago

Thanks a lot, it worked fine but it takes more time than np.count_nonzero

acdr · Accepted Answer · 2017-06-08 15:08:24Z

3

Your best bet is probably something like np.count_nonzero(x > threshold), where x is your 2-d array.

As the name implies, count_nonzero counts the number of elements that aren't zero. By making use of the fact that True is 1-ish, you can use it to count the number of elements that are True.

answered Jun 8, 2017 at 15:08

acdr

4,7863 gold badges24 silver badges48 bronze badges

Collectives™ on Stack Overflow

Conditional summation in python

2 Answers 2

4 Comments

Comments

Your Answer

Hot Network Questions

Collectives™ on Stack Overflow

2 Answers 2

4 Comments

Comments

Your Answer

Sign up or log in

Post as a guest

Related