mindspore.mint.baddbmm

mindspore.mint.baddbmm(input, batch1, batch2, *, beta=1, alpha=1)[source]

Perform a batch matrix-matrix product of matrices in batch1 and batch2 , input is added to the final result.

Note

batch1 and batch2 must be 3-D tensors each containing the same number of matrices.
When batch1 is a \((C, W, T)\) tensor and batch2 is a \((C, T, H)\) tensor, input must be broadcastable with \((C, W, H)\) tensor, and out will be a \((C, W, H)\) tensor.
If beta is 0, then input will be ignored.
beta and alpha must be integers when inputs of type not FloatTensor.

\[\text{out}_{i} = \beta \text{input}_{i} + \alpha (\text{batch1}_{i} \mathbin{@} \text{batch2}_{i})\]

Parameters

input (Tensor) – The input tensor.
batch1 (Tensor) – The first batch of matrices to be multiplied.
batch2 (Tensor) – The second batch of matrices to be multiplied.

Keyword Arguments

beta (Union[float, int], optional) – Scale factor for input. Default 1 .
alpha (Union[float, int], optional) – Scale factor for ( batch1 @ batch2 ). Default 1 .

Returns

Tensor

Supported Platforms:: Ascend

Examples

>>> import mindspore
>>> input = mindspore.mint.ones([3, 3])
>>> batch1 = mindspore.mint.arange(24.0).reshape((2, 3, 4))
>>> batch2 = mindspore.mint.arange(24.0).reshape((2, 4, 3))
>>> mindspore.mint.baddbmm(input, batch1, batch2)
Tensor(shape=[2, 3, 3], dtype=Float32, value=
[[[ 4.30000000e+01,  4.90000000e+01,  5.50000000e+01],
  [ 1.15000000e+02,  1.37000000e+02,  1.59000000e+02],
  [ 1.87000000e+02,  2.25000000e+02,  2.63000000e+02]],
 [[ 9.07000000e+02,  9.61000000e+02,  1.01500000e+03],
  [ 1.17100000e+03,  1.24100000e+03,  1.31100000e+03],
  [ 1.43500000e+03,  1.52100000e+03,  1.60700000e+03]]])