mindspore.Tensor.matmul

Tensor.matmul(tensor2) → Union[Tensor, numbers.Number]

Returns the matrix product of two tensors.

Note

Numpy arguments out, casting, order, subok, signature, and extobj are not supported.

The dtype of self and tensor2 must be same.

On Ascend platform, the dims of self and tensor2 must be between 1 and 6.
On GPU platform, the supported dtypes of self and tensor2 are ms.float16 and ms.float32.

Parameters

tensor2 (Tensor) – Input tensor, scalar not allowed. The last dimension of self must be the same size as the second last dimension of tensor2. And the shape of tensor and other could be broadcast.

Returns

Tensor or scalar, the matrix product of the inputs. This is a scalar only when both self and tensor2 are 1-d vectors.

Raises

TypeError – If the dtype of self and the dtype of tensor2 are not the same.
ValueError – If the last dimension of self is not the same size as the second-to-last dimension of tensor2, or if a scalar value is passed in.
ValueError – If the shape of self and tensor2 could not broadcast together.
RuntimeError – On Ascend platforms, the dims of self or tensor2 is less than 1 or greater than 6.

Supported Platforms:: Ascend GPU CPU

Examples

>>> import mindspore
>>> import numpy as np
>>> from mindspore import Tensor
>>> # case 1 : Reasonable application of broadcast mechanism
>>> input = Tensor(np.arange(2 * 3 * 4).reshape(2, 3, 4), mindspore.float32)
>>> other = Tensor(np.arange(4 * 5).reshape(4, 5), mindspore.float32)
>>> output = input.matmul(other)
>>> print(output)
[[[  70.   76.   82.   88.   94.]
  [ 190.  212.  234.  256.  278.]
  [ 310.  348.  386.  424.  462.]]
 [[ 430.  484.  538.  592.  646.]
  [ 550.  620.  690.  760.  830.]
  [ 670.  756.  842.  928. 1014.]]]
>>> print(output.shape)
(2, 3, 5)
>>> # case 2 : the rank of `tensor2` is 1
>>> input = Tensor(np.ones([1, 2]), mindspore.float32)
>>> other = Tensor(np.ones([2,]), mindspore.float32)
>>> output = input.matmul(other)
>>> print(output)
[2.]
>>> print(output.shape)
(1,)