std :: ml::gpu_mps
Function std::mps_sgd_update_f16_chained
int mps_sgd_update_f16_chained(
struct MpsBuf bufW,
struct MpsBuf bufGrad,
float lr,
unsigned long n,
)In-place SGD step against a persistent fp16/bf16 device buffer -- same shape as mps_sgd_update_f32_chained, weights never leave the GPU.