mindspore.Tensor.median

Tensor.median(axis=- 1, keepdims=False) → Tuple of Tensors

Computes the median and indices of input tensor.

Warning

indices does not necessarily contain the first occurrence of each median value found in the input, unless it is unique. The specific implementation of this API is device-specific. The results may be different on CPU and GPU.

Parameters

axis (int, optional) – Specify the axis for calculation. Default: -1 .
keepdims (bool, optional) – Whether the output tensor need to retain axis dimension or not. Default: False .

Returns

y (Tensor) - Returns the median value along the specified dimension. And It has the same dtype as the input.
indices (Tensor) - The index of the median. And the dtype is int64.

Raises

TypeError – If axis is not an int.
TypeError – If keepdims is not a bool.
ValueError – If axis is not in range of [-x.dim, x.dim-1].

Supported Platforms:: GPU CPU

Examples

>>> import numpy as np
>>> from mindspore import Tensor
>>> x = Tensor(np.array([[0.57, 0.11, 0.21],[0.38, 0.50, 0.57], [0.36, 0.16, 0.44]]).astype(np.float32))
>>> y = x.median(axis=0, keepdims=False)
>>> print(y)
(Tensor(shape=[3], dtype=Float32, value= [ 3.79999995e-01,  1.59999996e-01,  4.39999998e-01]),
Tensor(shape=[3], dtype=Int64, value= [1, 2, 2]))

Tensor.median() → Tensor

Return the median of the input.

Returns

y (Tensor) - Output median.

Supported Platforms:: Ascend

Tensor.median(dim=- 1, keepdim=False) → Tuple of Tensors

Output the median on the specified dimension dim and its corresponding index. If dim is None, calculate the median of all elements in the Tensor.

Parameters

dim (int, optional) – Specify the axis for calculation. Default: None .
keepdim (bool, optional) – Whether the output tensor need to retain dim dimension or not. Default: False .

Returns

y (Tensor) - Output median, with the same data type as input .
- If dim is None , y only has one element.
- If keepdim is True , the y has the same shape as the input except the shape of y in dimension dim is size 1.
- Otherwise, the y lacks dim dimension than input.
indices (Tensor) - The index of the median. Shape is consistent with y , with a data type of int64.

Raises

TypeError – If dim is not an int.
TypeError – If keepdim is not a bool.
ValueError – If dim is not in range of [-x.dim, x.dim-1].

Supported Platforms:: Ascend

Examples

>>> import numpy as np
>>> from mindspore import Tensor
>>> x = Tensor(np.array([[0.57, 0.11, 0.21],[0.38, 0.50, 0.57], [0.36, 0.16, 0.44]]).astype(np.float32))
>>> y = x.median(dim=0, keepdim=False)
>>> print(y)
(Tensor(shape=[3], dtype=Float32, value= [ 3.79999995e-01,  1.59999996e-01,  4.39999998e-01]),
Tensor(shape=[3], dtype=Int64, value= [1, 2, 2]))