torch.addbmm#

torch.addbmm(input,batch1,batch2,*,beta=1,alpha=1,out=None)→Tensor#

Performs a batch matrix-matrix product of matrices storedinbatch1 andbatch2,with a reduced add step (all matrix multiplications get accumulatedalong the first dimension).input is added to the final result.

batch1 andbatch2 must be 3-D tensors each containing thesame number of matrices.

Ifbatch1 is a$(b\times n\times m)$(b×n×m) tensor,batch2 is a$(b\times m\times p)$(b×m×p) tensor,input must bebroadcastable with a$(n\times p)$(n×p) tensorandout will be a$(n\times p)$(n×p) tensor.

$$out=\beta \text{ input}+\alpha (\sum _{i=0}^{b-1}{\text{batch1}}_i\mathrm{@}{\text{batch2}}_i)$$out=β input+α (i=0∑b−1​batch1i​@batch2i​)

Ifbeta is 0, then the content ofinput will be ignored, andnan andinf init will not be propagated.

For inputs of typeFloatTensor orDoubleTensor, argumentsbeta andalphamust be real numbers, otherwise they should be integers.

This operator supportsTensorFloat32.

On certain ROCm devices, when using float16 inputs this module will usedifferent precision for backward.

Parameters

• input (Tensor) – matrix to be added

• batch1 (Tensor) – the first batch of matrices to be multiplied

• batch2 (Tensor) – the second batch of matrices to be multiplied

Keyword Arguments

• beta (Number,optional) – multiplier forinput ($\beta$β)

• alpha (Number,optional) – multiplier forbatch1 @ batch2 ($\alpha$α)

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

Example:

>>>M=torch.randn(3,5)>>>batch1=torch.randn(10,3,4)>>>batch2=torch.randn(10,4,5)>>>torch.addbmm(M,batch1,batch2)tensor([[  6.6311,   0.0503,   6.9768, -12.0362,  -2.1653],        [ -4.8185,  -1.4255,  -6.6760,   8.9453,   2.5743],        [ -3.8202,   4.3691,   1.0943,  -1.1109,   5.4730]])