pub fn addmm(
c: &Array,
a: &Array,
b: &Array,
alpha: f32,
beta: f32,
) -> Result<Array>Expand description
alpha · (a @ b) + beta · c, with c broadcast to the product’s shape, as MLX’s
addmm. On the GPU it is one kernel (the GEMM’s epilogue), so Linear layers use it for
their bias.