std :: ml::gpu_rocm
Function std::rocm_flash_attention_f32
int rocm_flash_attention_f32(
const &float q,
const &float k,
const &float v,
&float out,
unsigned long heads,
unsigned long seqLen,
unsigned long kvLen,
unsigned long headDim,
float scale,
int causal,
)The attention ops above on host arrays (same layouts).