# torch.bmm¶

torch.bmm(input, mat2, *, out=None)Tensor

Performs a batch matrix-matrix product of matrices stored in input and mat2.

input and mat2 must be 3-D tensors each containing the same number of matrices.

If input is a $(b \times n \times m)$ tensor, mat2 is a $(b \times m \times p)$ tensor, out will be a $(b \times n \times p)$ tensor.

$\text{out}_i = \text{input}_i \mathbin{@} \text{mat2}_i$

This operator supports TensorFloat32.

Note

This function does not broadcast. For broadcasting matrix products, see torch.matmul().

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

• mat2 (Tensor) – the second batch of matrices to be multiplied

Keyword Arguments

out (Tensor, optional) – the output tensor.

Example:

>>> input = torch.randn(10, 3, 4)
>>> mat2 = torch.randn(10, 4, 5)
>>> res = torch.bmm(input, mat2)
>>> res.size()
torch.Size([10, 3, 5])