Python loader, IPython magics, backend discovery, and CLI launcher for Kiwi.
Project description
kiwi-array
kiwi-array is the Python loader package for Kiwi. It contains the
kiwi_array.bridge ctypes wrapper, IPython magics, runtime backend discovery,
and a small kiwi CLI launcher that execs the real native CLI from an
installed backend payload.
By default, installing kiwi-array also installs kiwi-array-host, the
conservative host backend. Accelerated backends are explicit extras:
kiwi-array[cpu], kiwi-array[metal], or kiwi-array[cuda12].
In IPython-compatible hosted notebooks, load the extension and use either the descriptive or short magic:
%load_ext kiwi_array
%%k
x:1 2 3
+/x
%%kiwi is registered as the explicit alias for %%k.
Use the lightweight diagnostics when sharing hosted notebook setup cells:
%kinfo
%ksmoke
On a GPU runtime, %ksmoke --gpu also pushes a small MLX-backed vector through
the native bridge and evaluates a gradient over it.
Runtime payload packages such as kiwi-array-host, kiwi-array-metal, and
kiwi-array-cuda12 register themselves through the kiwi_array.runtimes entry
point group. Set KIWI_RUNTIME=host, KIWI_RUNTIME=cpu,
KIWI_RUNTIME=metal, KIWI_RUNTIME=cuda12, or leave KIWI_RUNTIME=auto to
control runtime selection.
Vega-Lite outputs use the notebook's default light rendering by default. Set
KIWI_VEGALITE_THEME=dark before rendering if you want the HTML fallback to use
Vega-Embed's dark theme.
Project details
Release history Release notifications | RSS feed
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distributions
Built Distribution
Filter files by name, interpreter, ABI, and platform.
If you're not sure about the file name format, learn more about wheel file names.
Copy a direct link to the current filters
File details
Details for the file kiwi_array-0.2.49-py3-none-any.whl.
File metadata
- Download URL: kiwi_array-0.2.49-py3-none-any.whl
- Upload date:
- Size: 11.9 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/6.2.0 CPython/3.14.2
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 |
22b44f6fbaabd1d985cbfaba318cb8964ee594c61ddd36151c84488e48ff81ed
|
|
| MD5 |
22aecc6ca0a6ca9035215c624fb9cb44
|
|
| BLAKE2b-256 |
42c8536340f7b6414b1fdb1dbb8f7392d9d207973dd1bcf051b21d9f1235ce5c
|