chenyuandGitHub a666450e4d UOp pattern x + x -> x * 2 (#6224) ...

* UOp pattern x + x -> x * 2

now there's no NEG, with this it covers all kinds of a*x+b*x

* can remove x-x

2024-08-21 12:06:19 -04:00

..

codegen

UOp pattern x + x -> x * 2 (#6224)

2024-08-21 12:06:19 -04:00

engine

CIFAR trainer + various bugfixes / improvements (#6146)

2024-08-20 16:58:46 -07:00

nn

Fix track_running_stats in batchnorm (#6200)

2024-08-20 14:01:22 -07:00

renderer

Refactor amd kernel prefix (#6205)

2024-08-20 10:37:36 -07:00

runtime

nv fix size in SET_SEMAPHORE_A (#6213)

2024-08-21 01:47:10 +03:00

shape

re:6125 switch real_size to use uops [run_process_replay] (#6138)

2024-08-19 13:20:24 -04:00

__init__.py

add mnist tutorial

2024-04-25 16:08:32 +08:00

device.py

RUF018 assignment-in-assert [run_process_replay] (#6172)

2024-08-19 00:34:52 -04:00

dtype.py

threefry touchup [run_process_replay] (#6169)

2024-08-18 23:01:24 -04:00

function.py

remove UnaryOps.NEG from function.py (#6187)

2024-08-19 17:39:15 -04:00

helpers.py

Support gunzip in fetch (#6176)

2024-08-19 12:04:40 -07:00

lazy.py

Revert "delete arg from cast [run_process_replay] (#6202)" (#6214)

2024-08-20 16:45:30 -07:00

multi.py

MLB support reshape for uneven shards (#5804)

2024-08-01 02:36:03 -07:00

ops.py

no UnaryOps.NEG in generated UOp patterns (#6209)

2024-08-21 11:08:22 -04:00

py.typed

add a single py.typed (#6083)

2024-08-14 17:31:46 -07:00

tensor.py

remove UnaryOps.NEG from function.py (#6187)

2024-08-19 17:39:15 -04:00