oneflow.addmm

oneflow.addmm(beta=1, input, alpha=1, mat1, mat2, out=None) → Tensor

Performs a matrix multiplication of the matrices mat1 and mat2. The matrix input is added to the final result.

If mat1 is a \((n \times m)\) tensor, mat2 is a \((m \times p)\) tensor, then input must be broadcastable with a \((n \times p)\) tensor and out will be a \((n \times p)\) tensor.

alpha and beta are scaling factors on matrix-vector product between mat1 and mat2 and the added matrix input respectively.

\[\text{out} = \beta\ \text{input} + \alpha\ (\text{mat1}_i \mathbin{@} \text{mat2}_i)\]

For inputs of type float or double, arguments beta and alpha must be real numbers, otherwise they should be integers.

Parameters

  • beta (Number, optional) – multiplier for input (\(\beta\))

  • input (Tensor) – matrix to be added

  • alpha (Number, optional) – multiplier for \(mat1 @ mat2\) (\(\alpha\))

  • mat1 (Tensor) – the first matrix to be multiplied

  • mat2 (Tensor) – the second matrix to be multiplied

  • out (Tensor, optional) – the output tensor.

For example:

>>> import numpy as np
>>> import oneflow as flow
>>> input = flow.tensor(np.array([[1,2,4],[5,11,9.1]]))
>>> mat1 = flow.tensor(np.array([[7.3,1.9,7.3],[10.2,1,5.5]]))
>>> mat2 = flow.tensor(np.array([[7.3,1.9,7.3],[10.2,1,5.5],[3.7,2.2,8.1]]))
>>> output = flow.addmm(input, mat1, mat2)
>>> output
tensor([[100.6800,  33.8300, 126.8700],
        [110.0100,  43.4800, 133.6100]], dtype=oneflow.float64)
>>> output.shape
oneflow.Size([2, 3])

>>> input2 = flow.tensor(np.array([1.7]))
>>> mat1 = flow.tensor(np.array([[1,2],[5,9.1],[7.7,1.4]]))
>>> mat2 = flow.tensor(np.array([[1,2,3.7],[5,9.1,6.8]]))
>>> output2 = flow.addmm(input2, mat1, mat2, alpha=1, beta=2)
>>> output2
tensor([[14.4000, 23.6000, 20.7000],
        [53.9000, 96.2100, 83.7800],
        [18.1000, 31.5400, 41.4100]], dtype=oneflow.float64)
>>> output2.shape
oneflow.Size([3, 3])