5 Commits

Author	SHA1	Message	Date
Illia Silin	f46a6ffad8	Fix the fp8 gemm for large tensors on MI300. (#1011) ... * Fix the fp8 conversion * Try clipping value before conversion * Fix return * Simplify with a const * reduce the gemm input tensor values to reduce round-off error * replace if-else with lambda * fix syntax --------- Co-authored-by: Rostyslav Geyyer <rosty.geyyer@amd.com>	2023-10-27 21:10:47 -07:00
rocking	3696fe1c76	Layernorm and groupnorm support to save mean and inverse std in forward (#929) ... * save mean and inverse std in normalization * Save mean and inverse std in splitK * Vector save mean and inv std * Modify instance for save mean and std * simplify the layernorm example * Save mean and std in groupnorm example * Save mean and inv std in ckProfiler and test * Remove compute data type from base class * Save mean and inv std in client example * Add changelog * clang format * Fix compile error * Refine naming * Avoid error in bf16 * revert changelog	2023-10-19 07:36:29 +08:00
Illia Silin	b94fd0b227	update copyright headers (#726)	2023-05-31 18:46:57 -05:00
rocking	a1e344b1ae	Normalization/split k (#615)	2023-05-11 07:15:02 -05:00
rocking5566	ed3a2e5226	Groupnorm + swish external api (#668) ... * Rename to proper naming * Add example of groupnorm + swish * Extract duplicate code in example * Add groupnorm + swish instances * Ractor instance generation, split into multiple cpp file * Add external api and client example * Refine profiler message * Use ck math version of exp * Refine problem size in example * Add host version of exp	2023-04-10 08:02:17 -05:00