youngkingdom/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity
Files
dceabd4c5a2aa8cb29ce5a05311a57519baadddc
cutlass/include/cutlass
History
masahi dceabd4c5a Support half precision sigmoid activation (#378) 
* Support half precision sigmoid activation

* introduce a vectorized variant using fast_tanh

* move the math to fast_math.h

* fixed compile

* .raw() -> .to_half()

Co-authored-by: Haicheng Wu <haichengw@nvidia.com>
2021-12-22 14:45:06 -05:00
..
arch
Updates to fused epilogue (#383)
2021-12-17 16:04:43 -05:00
conv
CUTLASS 2.8 (#363)
2021-11-19 13:26:35 -08:00
epilogue
Support half precision sigmoid activation (#378)
2021-12-22 14:45:06 -05:00
gemm
Added missing synchronization to avoid WAR hazards between tiles. (#386)
2021-12-20 08:34:08 -08:00
layout
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
platform
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
reduction
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
thread
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
transform
CUTLASS 2.8 (#363)
2021-11-19 13:26:35 -08:00
aligned_buffer.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
array_planar_complex.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
array_subbyte.h
CUTLASS 2.8 (#363)
2021-11-19 13:26:35 -08:00
array.h
CUTLASS 2.8 (#363)
2021-11-19 13:26:35 -08:00
bfloat16.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
complex.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
constants.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
coord.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
core_io.h
CUTLASS 2.8 (#363)
2021-11-19 13:26:35 -08:00
cutlass.h
Updates to fused epilogue (#383)
2021-12-17 16:04:43 -05:00
device_kernel.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
fast_math.h
Support half precision sigmoid activation (#378)
2021-12-22 14:45:06 -05:00
functional.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
half.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
integer_subbyte.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
kernel_launch.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
matrix_coord.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
matrix_shape.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
matrix.h
CUTLASS 2.6.1 - functional and performance enhancements to strided DGRAD, fixes, and tuning
2021-09-03 10:26:15 -07:00
numeric_conversion.h
CUTLASS 2.8 (#363)
2021-11-19 13:26:35 -08:00
numeric_types.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
pitch_linear_coord.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
predicate_vector.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
quaternion.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
real.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
relatively_equal.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
semaphore.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
subbyte_reference.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
tensor_coord.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
tensor_ref_planar_complex.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
tensor_ref.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
tensor_view_planar_complex.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
tensor_view.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
tfloat32.h
Updates to fused epilogue (#383)
2021-12-17 16:04:43 -05:00
trace.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
uint128.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00
wmma_array.h
Cutlass 2.6 Update 1 (#301)
2021-07-27 17:58:30 -07:00