George Hotz fe71282ba1 faster RDNA assembly backend (#990) ...

* fast asm

* torch gemm

2023-06-16 12:06:38 -07:00

..

disk

Fast DiskTensor to other Tensor (#916)

2023-06-03 12:25:41 -07:00

gemm

faster RDNA assembly backend (#990)

2023-06-16 12:06:38 -07:00

intel

intel benchmark matmul gets 60 TFLOPS?

2023-06-04 17:01:50 +00:00

ptx

ptx example

2023-05-26 19:28:51 -07:00

rocm

faster RDNA assembly backend (#990)

2023-06-16 12:06:38 -07:00

triton

A Triton backend for tinygrad (#470)

2023-02-01 11:53:57 -08:00

archprobe.py

no werror in archprobe

2023-05-03 19:34:17 +00:00

augment.py

[Fix] fix typo: test_mnist -> datasets (#492)

2023-01-29 21:30:47 -08:00

gradcheck.py

Remove Tensor.data (#565)

2023-02-18 16:36:12 -08:00

helpers.py

RDNA assembly backend ($1000 bounty) (#787)

2023-06-16 09:33:18 -07:00

introspection.py

Devicebufferless (#708)

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

kernel_search.py

move to shapetracker.py

2023-03-11 07:50:07 -08:00

lib_test_ast.py

fix ast test

2023-03-04 07:49:25 -08:00

lr_scheduler.py

LR Schedulers (#755)

2023-05-27 07:47:49 -07:00

onnx_ops.py

One hot (#972)

2023-06-12 10:13:29 -07:00

onnx.py

fixes to Onnx ops LayerNormalization/Prelu and added OptionalHasElement/OptionalGetElement (#956)

2023-06-08 16:09:19 -07:00

thneed.py

multi cl_queue (#762)

2023-05-03 12:15:28 -07:00

training.py

add int64 as supported dtype from numpy (#699)

2023-03-18 17:15:04 -07:00

utils.py

Refactor LoadOps (#910)

2023-06-03 09:40:43 -07:00