Repository navigation
Cast input data as np.float64 in biweight functions - #6905
Conversation
|
Hi there @larrybradley 👋 - thanks for the pull request! I'm just a friendly 🤖 that checks for issues related to the changelog and making sure that this pull request is milestoned and labeled correctly. This is mainly intended for the maintainers, so if you are not a maintainer you can ignore this, and a maintainer will let you know if any action is required on your part 😃. Everything looks good from my point of view! 👍 If there are any issues with this message, please report them here. |
|
IMHO fixing underflow/overflow that happens often warrants a change log. Doesn't hurt, right? |
|
@pllim Done! |
|
Is adding a test possible? It seems like this fix is not currently tested in CI. 😅 |
|
@larrybradley - Could you add a test for this? and @crawfordsm could you sign it off, so it actually can go into 2.0.3? |
|
Test added! |
Cast input data as np.float64 in biweight functions
I often see floating underflows/overflows in the biweight stat functions when using float32 arrays. This PR casts the input arrays as float64, which prevents the runtime warnings.
This is more a fix than a feature, so I set the milestone as 2.0.3 -- feel free to change. Also, is a changelog necessary for this small change?