mlx/python/src
Awni Hannun 5798256fcf
Shapeless compilation for some graphs (#687)
* shapeless compilation for some graphs

* update compile benchmark

* default compile a few activations

* buffer donation

* bugfix

* shapeless fix

* update tests to work for cpu and gpu fusion

* test kwargs

* add kwargs to compile

* Recompile when python arguments change

* no compile for tanh

* some constant tests

---------

Co-authored-by: Angelos Katharopoulos <a_katharopoulos@apple.com>
2024-02-19 21:43:54 -08:00
..
array.cpp fix tolist for half types (#702) 2024-02-19 09:44:27 -08:00
CMakeLists.txt Adds device context manager (#679) 2024-02-14 14:14:58 -08:00
constants.cpp feat: Add numpy constants (#428) 2024-01-11 06:47:29 -08:00
device.cpp Adds device context manager (#679) 2024-02-14 14:14:58 -08:00
fast.cpp Custom primitive + RoPE fat op (#676) 2024-02-14 14:04:25 -08:00
fft.cpp copyright + ack 2023-11-30 11:12:53 -08:00
indexing.cpp Scatter vjp (#394) 2024-01-09 13:36:51 -08:00
indexing.h Scatter vjp (#394) 2024-01-09 13:36:51 -08:00
linalg.cpp QR factorization (#310) 2024-01-26 09:27:31 -08:00
load.cpp Metadata support for safetensors (#639) 2024-02-08 19:33:15 -08:00
load.h Metadata support for safetensors (#639) 2024-02-08 19:33:15 -08:00
metal.cpp Support disable metal buffer cache to prevent performance degradation caused by large memory caching (#390) 2024-01-18 08:33:34 -08:00
mlx.cpp Adds device context manager (#679) 2024-02-14 14:14:58 -08:00
ops.cpp Added support for atleast_1d, atleast_2d, atleast_3d (#694) 2024-02-19 09:40:52 -08:00
random.cpp Custom primitive + RoPE fat op (#676) 2024-02-14 14:04:25 -08:00
stream.cpp Adds device context manager (#679) 2024-02-14 14:14:58 -08:00
transforms.cpp Shapeless compilation for some graphs (#687) 2024-02-19 21:43:54 -08:00
utils.cpp Adds device context manager (#679) 2024-02-14 14:14:58 -08:00
utils.h Mlx array accessor (#128) 2023-12-11 13:42:55 -08:00