CUDA: Optimize PAD_REFLECT_1D #15957
CUDA: Optimize PAD_REFLECT_1D
1e29fafa
use fast_div to improve performance
9494833b
Apply suggestion from @JohannesGaessler
85835527
Apply suggestion from @JohannesGaessler
a5ef1d09
optimize
b3cf133a
use a concise expression to further speedup the cuda kernel
d73ba84a
add comment for rel_i0
e280cb87
Merge branch 'ggml-org:master' into PAD_REFLECT_1D_expriment
188ce93e
Merge branch 'ggml-org:master' into PAD_REFLECT_1D_expriment
4286ea78
Merge branch 'ggml-org:master' into PAD_REFLECT_1D_expriment
dd6789b1
Merge branch 'ggml-org:master' into PAD_REFLECT_1D_expriment
aa12620c
Assignees
No one assigned
Labels
testing
Nvidia GPU
ggml
Login to write a write a comment.
Login via GitHub