chenyuandGitHub 12f28ac9d4 catch runtime error in search._time_program (#3106) ...

return inf if search encountered runtime errors.

2024-01-12 21:53:13 -05:00

..

codegen

wmma: enable METAL half tensor cores and clean up cstyle (#3095)

2024-01-12 16:25:28 -05:00

features

catch runtime error in search._time_program (#3106)

2024-01-12 21:53:13 -05:00

nn

make Embedding device aware for multigpu (#3051)

2024-01-08 20:09:26 -08:00

renderer

wmma: enable METAL half tensor cores and clean up cstyle (#3095)

2024-01-12 16:25:28 -05:00

runtime

use device from LinearizerOptions in kernel search (#3090)

2024-01-11 14:46:03 -05:00

shape

update NumNode.__hash__ to be hash(self.b) (#3105)

2024-01-12 19:46:21 -05:00

__init__.py

move dtypes to dtype.py (#2964)

2024-01-01 14:58:48 -08:00

device.py

use device from LinearizerOptions in kernel search (#3090)

2024-01-11 14:46:03 -05:00

dtype.py

pass tuple for strs to startswith (#2986)

2024-01-02 19:51:15 -05:00

graph.py

early gate the graph (#3070)

2024-01-09 20:17:13 -08:00

helpers.py

mem_estimate is always int, not symbolic (#3083)

2024-01-10 23:39:51 -05:00

jit.py

update jit type annotation post lazy rewrite (#3091)

2024-01-11 15:49:30 -05:00

lazy.py

bugfix do not reset shapetracker of 0 size lazybuffer (#3096)

2024-01-11 23:22:52 -05:00

mlops.py

grad -> grad_output in mlops for consistency (#2982)

2024-01-02 18:03:55 -05:00

ops.py

remove unused reciprocal (#3053)

2024-01-09 08:59:04 -08:00

realize.py

mem_estimate is always int, not symbolic (#3083)

2024-01-10 23:39:51 -05:00

tensor.py

wmma: enable METAL half tensor cores and clean up cstyle (#3095)

2024-01-12 16:25:28 -05:00