torch.addr#

torch.addr(input,vec1,vec2,*,beta=1,alpha=1,out=None)→Tensor#

Performs the outer-product of vectorsvec1 andvec2and adds it to the matrixinput.

Optional valuesbeta andalpha are scaling factors on theouter product betweenvec1 andvec2 and the added matrixinput respectively.

$$\text{out}=\beta \text{ input}+\alpha (\text{vec1}\otimes \text{vec2})$$out=β input+α (vec1⊗vec2)

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

Ifvec1 is a vector of sizen andvec2 is a vectorof sizem, theninput must bebroadcastable with a matrix of size$(n\times m)$(n×m) andout will be a matrix of size$(n\times m)$(n×m).

Parameters

• input (Tensor) – matrix to be added

• vec1 (Tensor) – the first vector of the outer product

• vec2 (Tensor) – the second vector of the outer product

Keyword Arguments

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

• alpha (Number,optional) – multiplier for$\text{vec1}\otimes \text{vec2}$vec1⊗vec2 ($\alpha$α)

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

Example:

>>>vec1=torch.arange(1.,4.)>>>vec2=torch.arange(1.,3.)>>>M=torch.zeros(3,2)>>>torch.addr(M,vec1,vec2)tensor([[ 1.,  2.],        [ 2.,  4.],        [ 3.,  6.]])