huggingface/candle
1
0
Fork 0
mirror of https://github.com/huggingface/candle.git synced 2025-06-17 19:18:50 +00:00
Code Issues Packages Projects Releases Wiki Activity

Cuda implementation for copying data around.

Browse Source
This commit is contained in:
laurent
2023-06-23 11:18:29 +01:00
parent 79e4b29c2f
commit 1a90f9d3a6
5 changed files with 56 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
kernels/src/unary.cu
View File

@ -18,11 +18,14 @@ extern "C" __global__ void FN_NAME( \
} \
#if __CUDA_ARCH__ >= 530
UNARY_OP(__half, ucopy_f16, x)
UNARY_OP(__half, uneg_f16, -x)
UNARY_OP(__half, usqr_f16, x*x)
UNARY_OP(__half, usqrt_f16, sqrtg(x))
#endif
UNARY_OP(float, ucopy_f32, x)
UNARY_OP(float, ucopy_f64, x)
UNARY_OP(float, uneg_f32, -x)
UNARY_OP(float, uneg_f64, -x)
UNARY_OP(float, usqr_f32, x*x)