convert numpy warnings to errors

Question

I am getting

C:\Anaconda\lib\site-packages\numpy\core\_methods.py:59: RuntimeWarning: Mean of empty slice.
    warnings.warn("Mean of empty slice.", RuntimeWarning)

when I run a code I wrote. This is a large codebase. The problem is, I do not see where in my code this warning was fired.

I would rather have this as an error (e.g. an exception), so that I can see where this was generated in my code.

Is it possible to have exceptions instead of warnings?

Note: I solved my problem debugging line by line. But the question still stands.

score 15 · Answer 1 · answered Dec 09 '15 at 13:13

15

Setting

numpy.seterr(all='raise')

should do the trick. You should probably check out the details though.

answered Dec 09 '15 at 13:13

fbence

2,025
2
19
42

It did not. I found the offending line by debugging step by step. – O. Altun Dec 09 '15 at 13:21
1

No, this is the standard library's `warnings` mechanism, not the `seterr()` mechanism. – Robert Kern Dec 09 '15 at 14:18
my bad, should I delete the answer? – fbence Dec 09 '15 at 15:14
3

Worked for me with divide-by-zero warnings though. – Ray Jun 19 '19 at 14:44

score 5 · Answer 2 · answered Dec 09 '15 at 14:21

5

import warnings

warnings.filterwarnings('error', "Mean of empty slice.")

# ... Your code here.

See the documentation of the warnings module for more details.

answered Dec 09 '15 at 14:21

Robert Kern

13,118
3
35
32

convert numpy warnings to errors

2 Answers2

Linked