Thank you!

fixed

The Rust kata has a small bug, the initial function signature is

fn find_even_index(arr: &[i32]) -> usize

But the tests expect it to be

fn find_even_index(arr: &[i32]) -> Option<usize>

Fixable by the user but should be fixed in the kata

, I have the same problem, and don't know what is the problem...

i am having the very same problem, did you figure it out?

New challenges help us to increase our knowledge and make better algorithms. Congrats 👏
Don't forget to upvote if you liked it.
😁

Haha I just saw all the comments already! Well, it's the gift of reading lots of comments/questions on katas I suppose ;)

Numpy is a great tool. Specially if you've got to deal to lot of calculations very often. But Python itself has an incredible advantage, its integer precision that remains intact with only three arithmetic operators, not division (unless you use integer division). With Numpy, many calculation follow the float precision road, not the integer one. Hope that this helps.

@benjaminzwhite. You've got a special gift. You had seen it before it happened. :)

Yeah, I tested some numpy.datatypes and I get always 762096994298999808, as I do with 'normal' python floats, which is afaik a hardware specific.

Using numpy.float64 I got 762096994298999936 for the same test, still not good enough

Have the same issue as ofersadan85. It seems that there a rounding issues with very large inputs. I guess I have to use a specific data type like numpy.float64

Incorrect Result: 762096994298999808 should equal 762096994298999815

Anyone know what i'm doing wrong?