# tensorplay.functional.bmm

Source: https://www.tensorplay.cn/docs/generated/tensorplay.functional.bmm.html

```python
tensorplay.functional.bmm(input, mat2, out_dtype=None, *, 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](/docs/notes/numerical_accuracy.html#id6).

On certain ROCm devices, when using float16 inputs this module will use [different precision](/docs/notes/numerical_accuracy.html#fp16-on-mi200) for backward.

> **Note**
>
> This function does not broadcast. For broadcasting matrix products, see tensorplay.matmul().

Parameters:

- input ([Tensor](/docs/generated/tensorplay.Tensor.html#tensorplay.Tensor)) – the first batch of matrices to be multiplied

- mat2 ([Tensor](/docs/generated/tensorplay.Tensor.html#tensorplay.Tensor)) – the second batch of matrices to be multiplied

- out_dtype ([dtype](/docs/generated/tensorplay.DType.html#tensorplay.DType), optional) – the dtype of the output tensor, Supported only on CUDA and for tensorplay.float32 given tensorplay.float16/tensorplay.bfloat16 input dtypes

Keyword Arguments:

out ([Tensor](/docs/generated/tensorplay.Tensor.html#tensorplay.Tensor), optional) – the output tensor.

Example:

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