pub fn eval_systolic_bf16(
weights: &[f64],
acts: &[f64],
m: usize,
k: usize,
n: usize,
) -> Vec<f64>Expand description
bf16 GEMM: operands rounded to bf16, products and accumulation in f32 (the standard bf16 MAC), returned as
f64. weights is k*n row-major, acts is m*k row-major.