mindspore.mint.eye

mindspore.mint.eye(n, m=None, dtype=None)[source]

Creates a tensor with ones on the diagonal and zeros in the rest.

Note

The data type of returned tensor can be float16, float32, int8, int16, int32, int64, uint8 or bool on Ascend platforms.

Parameters
  • n (int) – The number of rows of returned tensor. Constant value only.

  • m (int, optional) – The number of columns of returned tensor. Constant value only. Default: None , if None , the number of columns is as the same as n.

  • dtype (mindspore.dtype, optional) – MindSpore's dtype, the data type of the returned tensor. The data type can be bool or Number. Default: None , the data type of the returned tensor is mindspore.float32.

Returns

Tensor, a tensor with ones on the diagonal and the rest of elements are zero. The shape of output depends on the user's Inputs n and m. And the data type depends on Inputs dtype.

Raises
Supported Platforms:

Ascend

Examples

>>> import mindspore
>>> from mindspore import mint
>>> output = mint.eye(2, 2, mindspore.int32)
>>> print(output)
[[1 0]
 [0 1]]
>>> print(output.dtype)
Int32
>>> output = mint.eye(1, 2, mindspore.float32)
>>> print(output)
[[1. 0.]]
>>> print(output.dtype)
Float32
>>> output = mint.eye(2, dtype=mindspore.int32)
>>> print(output)
[[1 0]
 [0 1]]
>>> print(output.dtype)
Int32
>>> output = mint.eye(2)
>>> print(output)
[[1. 0.]
 [0. 1.]]
>>> print(output.dtype)
Float32