Evaluate with NaN

Question

I'd like to compare a signal with the help of np.all() with a reference signal. The reference can contain np.nan. At this NaN-position it does not matter if the signal fulfills the comparison condition. Following is a sample code. But this code does not work. I could compare each element within a for-loop but is there a smart way to do this comparison?

import numpy as np reference = np.array([np.nan, 1]) signal = np.array([2, 2]) print np.all(reference < signal)

David Zwicker · Accepted Answer · 2013-03-26 13:37:49Z

2

Limit the comparison to the non-nan values with a proper slice. You could also use np.finite instead of ~np.isnan

import numpy as np reference = np.array([np.nan, 1]) signal = np.array([2, 2]) idx = ~np.isnan(reference) print np.all(reference[idx] < signal[idx])

edited Mar 26, 2013 at 13:37

answered Mar 26, 2013 at 13:23

David Zwicker

24.4k8 gold badges68 silver badges80 bronze badges

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

2 Comments

wewa Over a year ago

Thank you, works great. But there is a small mistake in your code. Please change idx = ~np.isnan(signal) to idx = ~np.isnan(reference).

David Zwicker Over a year ago

Oh, I'm sorry, I guess I didn't read your post carefully enough. I'm glad it helped, though!

Collectives™ on Stack Overflow

Evaluate with NaN

1 Answer 1

2 Comments

Hot Network Questions

Collectives™ on Stack Overflow

1 Answer 1

2 Comments

Related