Shaojie WANG f622889a5c use old ctile to avoid conv2d fwd bias relu add compute error (#271) ...

[ROCm/composable_kernel commit: 1c5d06f270]

2022-06-02 14:06:42 -05:00

..

host_utility

Navi21 gemm (#197)

2022-05-24 12:19:27 -05:00

problem_transform

Unified implementation of 1d/2d/3d conv bwd-data. fp32/fp16/bfp16/int8 (#134)

2022-03-29 10:52:25 -05:00

tensor

Compile for gfx908 and gfx90a (#130)

2022-03-31 12:33:34 -05:00

tensor_description

Resolution of issue #153: Add compiler warning on comparing int and size_t (#212)

2022-05-09 15:06:49 -05:00

tensor_operation/gpu

use old ctile to avoid conv2d fwd bias relu add compute error (#271)

2022-06-02 14:06:42 -05:00

utility

Unify the naming of the math functions used by the host and kernel (#262)

2022-06-01 21:49:53 -05:00

config.hpp

Gemm reduce max (#209)

2022-05-19 21:56:56 -05:00

options.hpp

remove options.hpp.in (#240)

2022-05-20 14:40:12 -05:00

stream_config.hpp

Add host API (#220)

2022-05-12 09:21:01 -05:00