gemm¶

dragon.math.gemm(
  inputs,
  alpha=1,
  beta=1,
  transpose_a=False,
  transpose_b=False,
  **kwargs
)[source]¶

Compute the general matrix multiplication.

\[\text{out} = \alpha AB + \beta C \]

Parameters:

• inputs (Sequence[dragon.Tensor]) – The matrix \(A\), \(B\) and optional \(C\).
• alpha (float, optional, default=1) – The value to \(\alpha\).
• beta (float, optional, default=1) – The value to \(\beta\).
• transpose_a (bool, optional, default=False) – True to transpose \(A\) before computation.
• transpose_b (bool, optional, default=False) – True to transpose \(B\) before computation.

Returns:

dragon.Tensor – The output tensor.