mindspore.mint.nn.AvgPool3d

查看源文件
class mindspore.mint.nn.AvgPool3d(kernel_size, stride=None, padding=0, ceil_mode=False, count_include_pad=True, divisor_override=None)[源代码]

对输入张量应用三维平均池化,可视为三维输入平面的组合。

警告

这是一个实验性API,后续可能修改或删除。

更多参考详见 mindspore.mint.nn.functional.avg_pool3d()

支持平台:

Ascend

样例:

>>> import mindspore as ms
>>> pool = ms.mint.nn.AvgPool3d(kernel_size=3, stride=1)
>>> x = ms.ops.randn(1, 2, 4, 4, 5).astype(ms.float32)
>>> output = pool(x)
>>> print(output.shape)
(1, 2, 2, 2, 3)
>>> x1 = ms.ops.randn(6, 5, 7, 7, 5).astype(ms.float32)
>>> pool2 = ms.mint.nn.AvgPool3d(4, stride=2, padding=(2, 2, 1), divisor_override=10)
>>> output2 = pool2(x1)
>>> print(output2.shape)
(6, 5, 4, 4, 2)