Jacky Lee fafe8e9ce2 casting: support all backends and implement half (#726) ...

* casting: support all backends and implement half

* map torch types in ops_torch

* reuse type map for torch buffer

* inverse dict lookup

2023-03-24 09:58:03 -07:00

..

codegen

clean up display name

2023-03-22 18:32:05 -07:00

nn

fix binop, other tests failure (#723)

2023-03-22 18:15:07 -07:00

runtime

casting: support all backends and implement half (#726)

2023-03-24 09:58:03 -07:00

shape

linearizer! (#714)

2023-03-19 23:43:49 -07:00

graph.py

print gflops avg with DEBUG=2

2023-03-23 16:07:08 -07:00

helpers.py

Devicebufferless (#708)

2023-03-18 14:40:23 -07:00

jit.py

Devicebufferless (#708)

2023-03-18 14:40:23 -07:00

lazy.py

hotcast bug fix

2023-03-23 11:49:47 -07:00

mlops.py

Fix constant folding (#713)

2023-03-18 17:52:46 -07:00

ops.py

OPTLOCAL=2 makes stable diffusion a usable speed after the cache builds

2023-03-22 19:19:11 -07:00

tensor.py

casting: support all backends and implement half (#726)

2023-03-24 09:58:03 -07:00