array-api
array-api copied to clipboard
RFC document, tooling and other content related to the array API standard
`xp.linalg.vecdot` is documented as alias to `xp.vecdot`. However, `xp.linalg.vecdot` signature uses `axis=None` default, while `xp.vecdot` has `axis=-1` as the default.   They should be consistent.
Background -- Consider a probability distribution library such as [efax](https://github.com/NeilGirdhar/efax/tree/main/efax). To add Array API support, each probability distribution class will contain a number of parameters, and it makes sense that...
This RFC proposes the addition of a new API in the array API specification for computing the element-wise reciprocal. ## Overview Based on array comparison [data](https://github.com/data-apis/array-api-comparison/blob/763ee61e4ce8d85c5431eee28d20b46a74626699/signatures/mathematical-functions-elementwise/reciprocal.md), the API is available...
This RFC proposes a new addition to the array API specification for counting the number of "non-zero" (i.e., truthy) values in an array. ## Overview Based on array comparison [data](https://github.com/data-apis/array-api-comparison/blob/01d47bde30f19ea4b2afa54f4c58ad62ab56e778/signatures/statistics/count_nonzero.md),...
This PR - resolves https://github.com/data-apis/array-api/issues/794 by adding support for counting the number of non-zero values to the specification. - is consistent with statistical functions (such as `max`, `mean`, `sum`, etc)...
This PR - resolves https://github.com/data-apis/array-api/issues/790 by adding support for computing the reciprocal to the specification. - requires that special cases be handled as if implemented as `1.0 / x`. -...
I've - Added text to the end of the term definitions indicating that RFC 2119 is used. - Add a Sphinx extension that automatically bolds RFC 2119 words. However, note...
Currently the specification for `astype`, as added in https://github.com/data-apis/array-api/pull/290, specifies that `copy=False/True` wherein there is no room for a "copy never" option, and the default `copy=True` means that calls to...
The standard is not clear what should happen in `cumulative_sum` for 0-D inputs https://data-apis.org/array-api/latest/API_specification/generated/array_api.cumulative_sum.html#cumulative-sum Note that NumPy and PyTorch have different conventions here: ```py >>> import numpy as np >>>...
This PR - resolves https://github.com/data-apis/array-api/issues/598 by adding `cumulative_prod` to the specification for computing the cumulative product. - follows `cumulative_sum` in matching `axis`, `dtype`, and `include_initial` semantics. - as with `cumulative_sum`,...