mindspore.ops.adaptive_max_pool1d

mindspore.ops.adaptive_max_pool1d(input, output_size)[source]

Applies a 1D adaptive maximum pooling over an input Tensor which can be regarded as a composition of 1D input planes.

Typically, the input is of shape \((N, C, L_{in})\), adaptive_max_pool1d outputs regional maximum in the \(L_{in}\)-dimension. The output is of shape \((N, C, L_{out})\), where \(L_{out}\) is defined by output_size.

Note

  • \(L_{in}\) must be divisible by output_size.

  • Ascend platform only supports float16 type for input.

Parameters

  • input (Tensor) – Tensor of shape \((N, C, L_{in})\), with float16 or float32 data type.

  • output_size (int) – the target output size \(L_{out}\).

Returns

Tensor of shape \((N, C, L_{out})\), has the same type as input.

Raises

  • TypeError – If input is neither float16 nor float32.

  • TypeError – If output_size is not an int.

  • ValueError – If output_size is less than 1.

  • ValueError – If the last dimension of input is smaller than output_size.

  • ValueError – If the last dimension of input is not divisible by output_size.

  • ValueError – If length of shape of input is not equal to 3.

Supported Platforms:

Ascend GPU CPU

Examples

>>> import mindspore
>>> import numpy as np
>>> from mindspore import Tensor, ops
>>> input = Tensor(np.random.randint(0, 10, [1, 3, 6]), mindspore.float32)
>>> output = ops.adaptive_max_pool1d(input, output_size=2)
>>> print(output.shape)
(1, 3, 2)