3 people 7c078191ce Misc rewrite perf improvements (#6500) ...

* Make UOp a normal class and use __slots__

* Use __slots__ in UPat

* Cache dtypes.{min,max}

* Use faster iterables in ops.py

* extend is a lot faster than nested listcomp

Co-authored-by: Roelof van Dijk <[email protected]>

---------

Co-authored-by: Roelof van Dijk <[email protected]>

2024-09-13 11:31:50 +08:00

..

codegen

Misc rewrite perf improvements (#6500)

2024-09-13 11:31:50 +08:00

engine

folding for vectorized consts [run_process_replay] (#6498)

2024-09-12 17:29:37 +08:00

nn

minor resnet cleanups (#6382)

2024-09-06 12:50:21 +08:00

renderer

gep tuple [run_process_replay] (#6495)

2024-09-12 16:37:31 +08:00

runtime

gep tuple [run_process_replay] (#6495)

2024-09-12 16:37:31 +08:00

shape

add void type to uop (#6471)

2024-09-11 18:16:28 +08:00

__init__.py

add mnist tutorial

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

device.py

hcq check queue size before submit (#6481)

2024-09-11 23:13:13 +03:00

dtype.py

Misc rewrite perf improvements (#6500)

2024-09-13 11:31:50 +08:00

function.py

math trait [run_process_replay] (#6364)

2024-09-05 16:19:17 +08:00

helpers.py

Even faster enums (#6483)

2024-09-12 20:08:02 +08:00

lazy.py

math trait [run_process_replay] (#6364)

2024-09-05 16:19:17 +08:00

multi.py

math trait [run_process_replay] (#6364)

2024-09-05 16:19:17 +08:00

ops.py

Misc rewrite perf improvements (#6500)

2024-09-13 11:31:50 +08:00

py.typed

add a single py.typed (#6083)

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

tensor.py

delete seen from the scheduler api [run_process_replay] (#6427)

2024-09-09 16:26:34 +08:00