You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Included because neither NumPy's, CuPy's, nor PyTorch's array objects
132
+
Included because neither NumPy's, CuPy's, Dask's, nor PyTorch's array objects
133
133
include this method. For NumPy, this function effectively does nothing since
134
134
the only supported device is the CPU, but for CuPy, this method supports
135
135
CuPy CUDA
@@ -240,6 +240,30 @@ Unlike the other libraries supported here, JAX array API support is contained
240
240
entirely in the JAX library. The JAX array API support is tracked at
241
241
https://github.com/google/jax/issues/18353.
242
242
243
+
## Dask
244
+
245
+
If you're using dask with numpy, many of the same limitations that apply to numpy
246
+
will also apply to dask. Besides those differences, other limitations include missing
247
+
sort functionality (no `sort` or `argsort`), and limited support for the optional `linalg`
248
+
and `fft` extensions.
249
+
250
+
In particular, the `fft` namespace is not compliant with the array API spec. Any functions
251
+
that you find under the `fft` namespace are the original, unwrapped functions under [`dask.array.fft`](https://docs.dask.org/en/latest/array-api.html#fast-fourier-transforms), which may or may not be Array API compliant. Use at your own risk!
252
+
253
+
For `linalg`, several methods are missing, for example:
254
+
-`cross`
255
+
-`det`
256
+
-`eigh`
257
+
-`eigvalsh`
258
+
-`matrix_power`
259
+
-`pinv`
260
+
-`slogdet`
261
+
-`matrix_norm`
262
+
-`matrix_rank`
263
+
Other methods may only be partially implemented or return incorrect results at times.
264
+
265
+
The minimum supported Dask version is 2023.12.0.
266
+
243
267
## Vendoring
244
268
245
269
This library supports vendoring as an installation method. To vendor the
0 commit comments