mindspore.Tensor.argmax

查看源文件
mindspore.Tensor.argmax(axis=None, keepdims=False)

返回tensor在指定轴上的最大值索引。

参数:
  • axis (Union[int, None],可选) - 指定计算轴。如果为 None ,计算tensor中的所有元素。默认 None

  • keepdims (bool,可选) - 输出tensor是否保留维度。默认 False

返回:

Tensor

支持平台:

Ascend GPU CPU

样例:

>>> import mindspore
>>> x = mindspore.tensor([[9, 3, 4, 5],
...                       [5, 2, 7, 4],
...                       [8, 1, 3, 6]])
>>> # case 1: By default, compute the maximum of all elements.
>>> x.argmax()
Tensor(shape=[], dtype=Int64, value= 0)
>>>
>>> # case 2: Compute the maximum along axis 1.
>>> x.argmax(axis=1)
Tensor(shape=[3], dtype=Int64, value= [0, 2, 0])
>>>
>>> # case 3: If keepdims=True, the output shape will be same of that of the input.
>>> x.argmax(axis=1, keepdims=True)
Tensor(shape=[3, 1], dtype=Int64, value=
[[0],
 [2],
 [0]])
mindspore.Tensor.argmax(dim=None, keepdim=False)

返回tensor在指定轴上的最大值索引。

参数:
  • dim (Union[int, None],可选) - 指定计算维度。如果为 None ,计算tensor中的所有元素。默认 None

  • keepdim (bool,可选) - 输出tensor是否保留维度。默认 False

返回:

Tensor

支持平台:

Ascend

样例:

>>> import mindspore
>>> x = mindspore.tensor([[9, 3, 4, 5],
...                       [5, 2, 7, 4],
...                       [8, 1, 3, 6]])
>>> # case 1: By default, compute the maximum of all elements.
>>> x.argmax()
Tensor(shape=[], dtype=Int64, value= 0)
>>>
>>> # case 2: Compute the maximum along dim 1.
>>> x.argmax(dim=1)
Tensor(shape=[3], dtype=Int64, value= [0, 2, 0])
>>>
>>> # case 3: If keepdim=True, the output shape will be same of that of the input.
>>> x.argmax(dim=1, keepdim=True)
Tensor(shape=[3, 1], dtype=Int64, value=
[[0],
 [2],
 [0]])