DOC: Documentation related finfo refactors and new slot addition #29889
Conversation
|
@seberg this is the first draft, you can have a look. |
seberg
left a comment
seberg
left a comment
There was a problem hiding this comment.
I would prefer much shorter release notes, but I like them so much I am also fine with keeping them :)
array.rst is way too large, but I suspect that dtypes.rst is only about the NumPy builtin DTypes and the whole new (and old) API for user DTypes is in array.rst.
That might be worth splitting up, but I still supect this is better to move there.
Anyway, love the docs otherwise, mostly wondering if it might be better to put in the other file.
|
This is ready for the review |
seberg
left a comment
seberg
left a comment
There was a problem hiding this comment.
Thanks for the update and working on this!
If you don't mind (and don't update), I think I might just apply the tiny changes and remove that one section and put it in very soon.
(I'll just leave shortening the release notes to Chuck if he thinks they are long.)
Co-authored-by: Sebastian Berg <[email protected]>
Co-authored-by: Sebastian Berg <[email protected]>
Co-authored-by: Sebastian Berg <[email protected]>
|
Cool, made all the suggestions
|
| platforms it is IEEE 754 binary128 (quad precision) or binary64-extended | ||
| (80-bit extended precision). On PowerPC systems, it may use the IBM | ||
| double-double format (a pair of float64 values), which has special | ||
| characteristics for precision and range. |
seberg
changed the title
|
Thanks @SwayamInSync! |
…umpy#29836 (numpy#29889) This is a follow-up PR of work happened in numpygh-29836 * The deprecated MachAr runtime discovery mechanism has been removed. * np.finfo fetches the constants provided by the compiler macros * new slot to fetch the dtype related constants
2 participants
This is a follow-up PR of work happened in #29836
Summary of prev work
MachArruntime discovery mechanism has been removed.np.finfofetches the constants provided by the compiler macros