tinygrad

geohot/tinygrad

Fork 0

forked from tinygrad/tinygrad

Files

T

History

George HotzandGitHub 817b64f5e5 A conv is a reduce op (#356 )

* universal strided conv

* more correct

* hmm, CPU works

* cleaner cl code output

* make noconv a flag

* cleanup __getitem__

* refactor broadcasting

* put that back

* unneeded reshape in getitem

* fix strided for torch

2022-07-10 19:58:50 -07:00

ane

refactor accel

2021-10-30 16:10:59 -07:00

cherry

only resnet18, it's too slow otherwise

2021-10-30 16:48:39 -07:00

cuda

Accel/cuda (#319 )

2022-05-14 21:25:30 -07:00

lazy

reorg opts, nicer graph

2022-07-02 22:29:09 -07:00

metal

a bit of work on metal

2021-12-30 13:53:08 -05:00

opencl

A conv is a reduce op (#356 )

2022-07-10 19:58:50 -07:00

tpu

header

2021-10-30 16:41:05 -07:00

README

refactor efficientnet loading

2021-10-30 17:02:17 -07:00

README

This is where we scope out adding accelerators to tinygrad

ane -- Apple Neural Engine, in the M1 + newer iPhones
cherry -- Largely defunct custom hardware based on a RISC-V extension
tpu -- Google's TPU, available for rent in Google Cloud