mindspore.ops.ger

mindspore.ops.ger(input, vec2)[source]

Ger product of input and vec2. Calculate the outer product of two arrays. If input is a 1D Tensor of shape \((m,)\) and vec2 is a 1D Tensor of shape \((n,)\), then output must be a 2D Tensor of shape \((m, n)\).

Note

Currently Ascend does not support float64 data input.

Parameters

input (Tensor) – input Tensor, with dtype of float16, float32 or float64.
vec2 (Tensor) – input Tensor, with dtype of float16, float32 or float64, must have the same dtype as input.

Returns

Tensor, output matrix with the same dtype as inputs. With input shape \((m,)\) and vec2 shape of \((n,)\), the output has shape \((m, n)\).

Raises

TypeError – If input or vec2 is not a 1-D Tensor.
TypeError – If the dtype of input and vec2 is not float16, float32 or float64.
TypeError – If the dtype of input and vec2 are not the same.

Supported Platforms:: Ascend GPU CPU

Examples

>>> input = Tensor([1., 2., 3., 4.], mindspore.float32)
>>> vec2 = Tensor([1., 2., 3.], mindspore.float32)
>>> output = ops.ger(input, vec2)
>>> print(output)
[[ 1.  2.  3.]
 [ 2.  4.  6.]
 [ 3.  6.  9.]
 [ 4.  8. 12.]]