[docs]classMaxPool1d(_MaxPoolNd):r"""Applies a 1D max pooling over an input signal composed of several input planes. In the simplest case, the output value of the layer with input size :math:`(N, C, L)` and output :math:`(N, C, L_{out})` can be precisely described as: .. math:: out(N_i, C_j, k) = \max_{m=0, \ldots, \text{kernel\_size} - 1} input(N_i, C_j, stride \times k + m) If :attr:`padding` is non-zero, then the input is implicitly padded with negative infinity on both sides for :attr:`padding` number of points. :attr:`dilation` is the stride between the elements within the sliding window. This `link`_ has a nice visualization of the pooling parameters. Note: When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding or the input. Sliding windows that would start in the right padded region are ignored. Args: kernel_size: The size of the sliding window, must be > 0. stride: The stride of the sliding window, must be > 0. Default value is :attr:`kernel_size`. padding: Implicit negative infinity padding to be added on both sides, must be >= 0 and <= kernel_size / 2. dilation: The stride between elements within a sliding window, must be > 0. return_indices: If ``True``, will return the argmax along with the max values. Useful for :class:`torch.nn.MaxUnpool1d` later ceil_mode: If ``True``, will use `ceil` instead of `floor` to compute the output shape. This ensures that every element in the input tensor is covered by a sliding window. Shape: - Input: :math:`(N, C, L_{in})` or :math:`(C, L_{in})`. - Output: :math:`(N, C, L_{out})` or :math:`(C, L_{out})`, where .. math:: L_{out} = \left\lfloor \frac{L_{in} + 2 \times \text{padding} - \text{dilation} \times (\text{kernel\_size} - 1) - 1}{\text{stride}} + 1\right\rfloor Examples:: >>> # pool of size=3, stride=2 >>> m = nn.MaxPool1d(3, stride=2) >>> input = torch.randn(20, 16, 50) >>> output = m(input) .. _link: https://github.com/vdumoulin/conv_arithmetic/blob/master/README.md """kernel_size:_size_1_tstride:_size_1_tpadding:_size_1_tdilation:_size_1_tdefforward(self,input:Tensor):returnF.max_pool1d(input,self.kernel_size,self.stride,self.padding,self.dilation,ceil_mode=self.ceil_mode,return_indices=self.return_indices)
[docs]classMaxPool2d(_MaxPoolNd):r"""Applies a 2D max pooling over an input signal composed of several input planes. In the simplest case, the output value of the layer with input size :math:`(N, C, H, W)`, output :math:`(N, C, H_{out}, W_{out})` and :attr:`kernel_size` :math:`(kH, kW)` can be precisely described as: .. math:: \begin{aligned} out(N_i, C_j, h, w) ={} & \max_{m=0, \ldots, kH-1} \max_{n=0, \ldots, kW-1} \\ & \text{input}(N_i, C_j, \text{stride[0]} \times h + m, \text{stride[1]} \times w + n) \end{aligned} If :attr:`padding` is non-zero, then the input is implicitly padded with negative infinity on both sides for :attr:`padding` number of points. :attr:`dilation` controls the spacing between the kernel points. It is harder to describe, but this `link`_ has a nice visualization of what :attr:`dilation` does. Note: When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding or the input. Sliding windows that would start in the right padded region are ignored. The parameters :attr:`kernel_size`, :attr:`stride`, :attr:`padding`, :attr:`dilation` can either be: - a single ``int`` -- in which case the same value is used for the height and width dimension - a ``tuple`` of two ints -- in which case, the first `int` is used for the height dimension, and the second `int` for the width dimension Args: kernel_size: the size of the window to take a max over stride: the stride of the window. Default value is :attr:`kernel_size` padding: implicit zero padding to be added on both sides dilation: a parameter that controls the stride of elements in the window return_indices: if ``True``, will return the max indices along with the outputs. Useful for :class:`torch.nn.MaxUnpool2d` later ceil_mode: when True, will use `ceil` instead of `floor` to compute the output shape Shape: - Input: :math:`(N, C, H_{in}, W_{in})` or :math:`(C, H_{in}, W_{in})` - Output: :math:`(N, C, H_{out}, W_{out})` or :math:`(C, H_{out}, W_{out})`, where .. math:: H_{out} = \left\lfloor\frac{H_{in} + 2 * \text{padding[0]} - \text{dilation[0]} \times (\text{kernel\_size[0]} - 1) - 1}{\text{stride[0]}} + 1\right\rfloor .. math:: W_{out} = \left\lfloor\frac{W_{in} + 2 * \text{padding[1]} - \text{dilation[1]} \times (\text{kernel\_size[1]} - 1) - 1}{\text{stride[1]}} + 1\right\rfloor Examples:: >>> # pool of square window of size=3, stride=2 >>> m = nn.MaxPool2d(3, stride=2) >>> # pool of non-square window >>> m = nn.MaxPool2d((3, 2), stride=(2, 1)) >>> input = torch.randn(20, 16, 50, 32) >>> output = m(input) .. _link: https://github.com/vdumoulin/conv_arithmetic/blob/master/README.md """kernel_size:_size_2_tstride:_size_2_tpadding:_size_2_tdilation:_size_2_tdefforward(self,input:Tensor):returnF.max_pool2d(input,self.kernel_size,self.stride,self.padding,self.dilation,ceil_mode=self.ceil_mode,return_indices=self.return_indices)
[docs]classMaxPool3d(_MaxPoolNd):r"""Applies a 3D max pooling over an input signal composed of several input planes. In the simplest case, the output value of the layer with input size :math:`(N, C, D, H, W)`, output :math:`(N, C, D_{out}, H_{out}, W_{out})` and :attr:`kernel_size` :math:`(kD, kH, kW)` can be precisely described as: .. math:: \begin{aligned} \text{out}(N_i, C_j, d, h, w) ={} & \max_{k=0, \ldots, kD-1} \max_{m=0, \ldots, kH-1} \max_{n=0, \ldots, kW-1} \\ & \text{input}(N_i, C_j, \text{stride[0]} \times d + k, \text{stride[1]} \times h + m, \text{stride[2]} \times w + n) \end{aligned} If :attr:`padding` is non-zero, then the input is implicitly padded with negative infinity on both sides for :attr:`padding` number of points. :attr:`dilation` controls the spacing between the kernel points. It is harder to describe, but this `link`_ has a nice visualization of what :attr:`dilation` does. Note: When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding or the input. Sliding windows that would start in the right padded region are ignored. The parameters :attr:`kernel_size`, :attr:`stride`, :attr:`padding`, :attr:`dilation` can either be: - a single ``int`` -- in which case the same value is used for the depth, height and width dimension - a ``tuple`` of three ints -- in which case, the first `int` is used for the depth dimension, the second `int` for the height dimension and the third `int` for the width dimension Args: kernel_size: the size of the window to take a max over stride: the stride of the window. Default value is :attr:`kernel_size` padding: implicit zero padding to be added on all three sides dilation: a parameter that controls the stride of elements in the window return_indices: if ``True``, will return the max indices along with the outputs. Useful for :class:`torch.nn.MaxUnpool3d` later ceil_mode: when True, will use `ceil` instead of `floor` to compute the output shape Shape: - Input: :math:`(N, C, D_{in}, H_{in}, W_{in})` or :math:`(C, D_{in}, H_{in}, W_{in})`. - Output: :math:`(N, C, D_{out}, H_{out}, W_{out})` or :math:`(C, D_{out}, H_{out}, W_{out})`, where .. math:: D_{out} = \left\lfloor\frac{D_{in} + 2 \times \text{padding}[0] - \text{dilation}[0] \times (\text{kernel\_size}[0] - 1) - 1}{\text{stride}[0]} + 1\right\rfloor .. math:: H_{out} = \left\lfloor\frac{H_{in} + 2 \times \text{padding}[1] - \text{dilation}[1] \times (\text{kernel\_size}[1] - 1) - 1}{\text{stride}[1]} + 1\right\rfloor .. math:: W_{out} = \left\lfloor\frac{W_{in} + 2 \times \text{padding}[2] - \text{dilation}[2] \times (\text{kernel\_size}[2] - 1) - 1}{\text{stride}[2]} + 1\right\rfloor Examples:: >>> # pool of square window of size=3, stride=2 >>> m = nn.MaxPool3d(3, stride=2) >>> # pool of non-square window >>> m = nn.MaxPool3d((3, 2, 2), stride=(2, 1, 2)) >>> input = torch.randn(20, 16, 50,44, 31) >>> output = m(input) .. _link: https://github.com/vdumoulin/conv_arithmetic/blob/master/README.md """# noqa: E501kernel_size:_size_3_tstride:_size_3_tpadding:_size_3_tdilation:_size_3_tdefforward(self,input:Tensor):returnF.max_pool3d(input,self.kernel_size,self.stride,self.padding,self.dilation,ceil_mode=self.ceil_mode,return_indices=self.return_indices)
[docs]classMaxUnpool1d(_MaxUnpoolNd):r"""Computes a partial inverse of :class:`MaxPool1d`. :class:`MaxPool1d` is not fully invertible, since the non-maximal values are lost. :class:`MaxUnpool1d` takes in as input the output of :class:`MaxPool1d` including the indices of the maximal values and computes a partial inverse in which all non-maximal values are set to zero. .. note:: :class:`MaxPool1d` can map several input sizes to the same output sizes. Hence, the inversion process can get ambiguous. To accommodate this, you can provide the needed output size as an additional argument :attr:`output_size` in the forward call. See the Inputs and Example below. Args: kernel_size (int or tuple): Size of the max pooling window. stride (int or tuple): Stride of the max pooling window. It is set to :attr:`kernel_size` by default. padding (int or tuple): Padding that was added to the input Inputs: - `input`: the input Tensor to invert - `indices`: the indices given out by :class:`~torch.nn.MaxPool1d` - `output_size` (optional): the targeted output size Shape: - Input: :math:`(N, C, H_{in})` or :math:`(C, H_{in})`. - Output: :math:`(N, C, H_{out})` or :math:`(C, H_{out})`, where .. math:: H_{out} = (H_{in} - 1) \times \text{stride}[0] - 2 \times \text{padding}[0] + \text{kernel\_size}[0] or as given by :attr:`output_size` in the call operator Example:: >>> # xdoctest: +IGNORE_WANT("do other tests modify the global state?") >>> pool = nn.MaxPool1d(2, stride=2, return_indices=True) >>> unpool = nn.MaxUnpool1d(2, stride=2) >>> input = torch.tensor([[[1., 2, 3, 4, 5, 6, 7, 8]]]) >>> output, indices = pool(input) >>> unpool(output, indices) tensor([[[ 0., 2., 0., 4., 0., 6., 0., 8.]]]) >>> # Example showcasing the use of output_size >>> input = torch.tensor([[[1., 2, 3, 4, 5, 6, 7, 8, 9]]]) >>> output, indices = pool(input) >>> unpool(output, indices, output_size=input.size()) tensor([[[ 0., 2., 0., 4., 0., 6., 0., 8., 0.]]]) >>> unpool(output, indices) tensor([[[ 0., 2., 0., 4., 0., 6., 0., 8.]]]) """kernel_size:_size_1_tstride:_size_1_tpadding:_size_1_tdef__init__(self,kernel_size:_size_1_t,stride:Optional[_size_1_t]=None,padding:_size_1_t=0)->None:super(MaxUnpool1d,self).__init__()self.kernel_size=_single(kernel_size)self.stride=_single(strideif(strideisnotNone)elsekernel_size)self.padding=_single(padding)defforward(self,input:Tensor,indices:Tensor,output_size:Optional[List[int]]=None)->Tensor:returnF.max_unpool1d(input,indices,self.kernel_size,self.stride,self.padding,output_size)
[docs]classMaxUnpool2d(_MaxUnpoolNd):r"""Computes a partial inverse of :class:`MaxPool2d`. :class:`MaxPool2d` is not fully invertible, since the non-maximal values are lost. :class:`MaxUnpool2d` takes in as input the output of :class:`MaxPool2d` including the indices of the maximal values and computes a partial inverse in which all non-maximal values are set to zero. .. note:: :class:`MaxPool2d` can map several input sizes to the same output sizes. Hence, the inversion process can get ambiguous. To accommodate this, you can provide the needed output size as an additional argument :attr:`output_size` in the forward call. See the Inputs and Example below. Args: kernel_size (int or tuple): Size of the max pooling window. stride (int or tuple): Stride of the max pooling window. It is set to :attr:`kernel_size` by default. padding (int or tuple): Padding that was added to the input Inputs: - `input`: the input Tensor to invert - `indices`: the indices given out by :class:`~torch.nn.MaxPool2d` - `output_size` (optional): the targeted output size Shape: - Input: :math:`(N, C, H_{in}, W_{in})` or :math:`(C, H_{in}, W_{in})`. - Output: :math:`(N, C, H_{out}, W_{out})` or :math:`(C, H_{out}, W_{out})`, where .. math:: H_{out} = (H_{in} - 1) \times \text{stride[0]} - 2 \times \text{padding[0]} + \text{kernel\_size[0]} .. math:: W_{out} = (W_{in} - 1) \times \text{stride[1]} - 2 \times \text{padding[1]} + \text{kernel\_size[1]} or as given by :attr:`output_size` in the call operator Example:: >>> pool = nn.MaxPool2d(2, stride=2, return_indices=True) >>> unpool = nn.MaxUnpool2d(2, stride=2) >>> input = torch.tensor([[[[ 1., 2., 3., 4.], [ 5., 6., 7., 8.], [ 9., 10., 11., 12.], [13., 14., 15., 16.]]]]) >>> output, indices = pool(input) >>> unpool(output, indices) tensor([[[[ 0., 0., 0., 0.], [ 0., 6., 0., 8.], [ 0., 0., 0., 0.], [ 0., 14., 0., 16.]]]]) >>> # Now using output_size to resolve an ambiguous size for the inverse >>> input = torch.torch.tensor([[[[ 1., 2., 3., 4., 5.], [ 6., 7., 8., 9., 10.], [11., 12., 13., 14., 15.], [16., 17., 18., 19., 20.]]]]) >>> output, indices = pool(input) >>> # This call will not work without specifying output_size >>> unpool(output, indices, output_size=input.size()) tensor([[[[ 0., 0., 0., 0., 0.], [ 0., 7., 0., 9., 0.], [ 0., 0., 0., 0., 0.], [ 0., 17., 0., 19., 0.]]]]) """kernel_size:_size_2_tstride:_size_2_tpadding:_size_2_tdef__init__(self,kernel_size:_size_2_t,stride:Optional[_size_2_t]=None,padding:_size_2_t=0)->None:super(MaxUnpool2d,self).__init__()self.kernel_size=_pair(kernel_size)self.stride=_pair(strideif(strideisnotNone)elsekernel_size)self.padding=_pair(padding)defforward(self,input:Tensor,indices:Tensor,output_size:Optional[List[int]]=None)->Tensor:returnF.max_unpool2d(input,indices,self.kernel_size,self.stride,self.padding,output_size)
[docs]classMaxUnpool3d(_MaxUnpoolNd):r"""Computes a partial inverse of :class:`MaxPool3d`. :class:`MaxPool3d` is not fully invertible, since the non-maximal values are lost. :class:`MaxUnpool3d` takes in as input the output of :class:`MaxPool3d` including the indices of the maximal values and computes a partial inverse in which all non-maximal values are set to zero. .. note:: :class:`MaxPool3d` can map several input sizes to the same output sizes. Hence, the inversion process can get ambiguous. To accommodate this, you can provide the needed output size as an additional argument :attr:`output_size` in the forward call. See the Inputs section below. Args: kernel_size (int or tuple): Size of the max pooling window. stride (int or tuple): Stride of the max pooling window. It is set to :attr:`kernel_size` by default. padding (int or tuple): Padding that was added to the input Inputs: - `input`: the input Tensor to invert - `indices`: the indices given out by :class:`~torch.nn.MaxPool3d` - `output_size` (optional): the targeted output size Shape: - Input: :math:`(N, C, D_{in}, H_{in}, W_{in})` or :math:`(C, D_{in}, H_{in}, W_{in})`. - Output: :math:`(N, C, D_{out}, H_{out}, W_{out})` or :math:`(C, D_{out}, H_{out}, W_{out})`, where .. math:: D_{out} = (D_{in} - 1) \times \text{stride[0]} - 2 \times \text{padding[0]} + \text{kernel\_size[0]} .. math:: H_{out} = (H_{in} - 1) \times \text{stride[1]} - 2 \times \text{padding[1]} + \text{kernel\_size[1]} .. math:: W_{out} = (W_{in} - 1) \times \text{stride[2]} - 2 \times \text{padding[2]} + \text{kernel\_size[2]} or as given by :attr:`output_size` in the call operator Example:: >>> # pool of square window of size=3, stride=2 >>> pool = nn.MaxPool3d(3, stride=2, return_indices=True) >>> unpool = nn.MaxUnpool3d(3, stride=2) >>> output, indices = pool(torch.randn(20, 16, 51, 33, 15)) >>> unpooled_output = unpool(output, indices) >>> unpooled_output.size() torch.Size([20, 16, 51, 33, 15]) """kernel_size:_size_3_tstride:_size_3_tpadding:_size_3_tdef__init__(self,kernel_size:_size_3_t,stride:Optional[_size_3_t]=None,padding:_size_3_t=0)->None:super(MaxUnpool3d,self).__init__()self.kernel_size=_triple(kernel_size)self.stride=_triple(strideif(strideisnotNone)elsekernel_size)self.padding=_triple(padding)defforward(self,input:Tensor,indices:Tensor,output_size:Optional[List[int]]=None)->Tensor:returnF.max_unpool3d(input,indices,self.kernel_size,self.stride,self.padding,output_size)
[docs]classAvgPool1d(_AvgPoolNd):r"""Applies a 1D average pooling over an input signal composed of several input planes. In the simplest case, the output value of the layer with input size :math:`(N, C, L)`, output :math:`(N, C, L_{out})` and :attr:`kernel_size` :math:`k` can be precisely described as: .. math:: \text{out}(N_i, C_j, l) = \frac{1}{k} \sum_{m=0}^{k-1} \text{input}(N_i, C_j, \text{stride} \times l + m) If :attr:`padding` is non-zero, then the input is implicitly zero-padded on both sides for :attr:`padding` number of points. Note: When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding or the input. Sliding windows that would start in the right padded region are ignored. The parameters :attr:`kernel_size`, :attr:`stride`, :attr:`padding` can each be an ``int`` or a one-element tuple. Args: kernel_size: the size of the window stride: the stride of the window. Default value is :attr:`kernel_size` padding: implicit zero padding to be added on both sides ceil_mode: when True, will use `ceil` instead of `floor` to compute the output shape count_include_pad: when True, will include the zero-padding in the averaging calculation Shape: - Input: :math:`(N, C, L_{in})` or :math:`(C, L_{in})`. - Output: :math:`(N, C, L_{out})` or :math:`(C, L_{out})`, where .. math:: L_{out} = \left\lfloor \frac{L_{in} + 2 \times \text{padding} - \text{kernel\_size}}{\text{stride}} + 1\right\rfloor Examples:: >>> # pool with window of size=3, stride=2 >>> m = nn.AvgPool1d(3, stride=2) >>> m(torch.tensor([[[1.,2,3,4,5,6,7]]])) tensor([[[2., 4., 6.]]]) """kernel_size:_size_1_tstride:_size_1_tpadding:_size_1_tceil_mode:boolcount_include_pad:booldef__init__(self,kernel_size:_size_1_t,stride:_size_1_t=None,padding:_size_1_t=0,ceil_mode:bool=False,count_include_pad:bool=True)->None:super(AvgPool1d,self).__init__()self.kernel_size=_single(kernel_size)self.stride=_single(strideifstrideisnotNoneelsekernel_size)self.padding=_single(padding)self.ceil_mode=ceil_modeself.count_include_pad=count_include_paddefforward(self,input:Tensor)->Tensor:returnF.avg_pool1d(input,self.kernel_size,self.stride,self.padding,self.ceil_mode,self.count_include_pad)
[docs]classAvgPool2d(_AvgPoolNd):r"""Applies a 2D average pooling over an input signal composed of several input planes. In the simplest case, the output value of the layer with input size :math:`(N, C, H, W)`, output :math:`(N, C, H_{out}, W_{out})` and :attr:`kernel_size` :math:`(kH, kW)` can be precisely described as: .. math:: out(N_i, C_j, h, w) = \frac{1}{kH * kW} \sum_{m=0}^{kH-1} \sum_{n=0}^{kW-1} input(N_i, C_j, stride[0] \times h + m, stride[1] \times w + n) If :attr:`padding` is non-zero, then the input is implicitly zero-padded on both sides for :attr:`padding` number of points. Note: When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding or the input. Sliding windows that would start in the right padded region are ignored. The parameters :attr:`kernel_size`, :attr:`stride`, :attr:`padding` can either be: - a single ``int`` -- in which case the same value is used for the height and width dimension - a ``tuple`` of two ints -- in which case, the first `int` is used for the height dimension, and the second `int` for the width dimension Args: kernel_size: the size of the window stride: the stride of the window. Default value is :attr:`kernel_size` padding: implicit zero padding to be added on both sides ceil_mode: when True, will use `ceil` instead of `floor` to compute the output shape count_include_pad: when True, will include the zero-padding in the averaging calculation divisor_override: if specified, it will be used as divisor, otherwise size of the pooling region will be used. Shape: - Input: :math:`(N, C, H_{in}, W_{in})` or :math:`(C, H_{in}, W_{in})`. - Output: :math:`(N, C, H_{out}, W_{out})` or :math:`(C, H_{out}, W_{out})`, where .. math:: H_{out} = \left\lfloor\frac{H_{in} + 2 \times \text{padding}[0] - \text{kernel\_size}[0]}{\text{stride}[0]} + 1\right\rfloor .. math:: W_{out} = \left\lfloor\frac{W_{in} + 2 \times \text{padding}[1] - \text{kernel\_size}[1]}{\text{stride}[1]} + 1\right\rfloor Examples:: >>> # pool of square window of size=3, stride=2 >>> m = nn.AvgPool2d(3, stride=2) >>> # pool of non-square window >>> m = nn.AvgPool2d((3, 2), stride=(2, 1)) >>> input = torch.randn(20, 16, 50, 32) >>> output = m(input) """__constants__=['kernel_size','stride','padding','ceil_mode','count_include_pad','divisor_override']kernel_size:_size_2_tstride:_size_2_tpadding:_size_2_tceil_mode:boolcount_include_pad:booldef__init__(self,kernel_size:_size_2_t,stride:Optional[_size_2_t]=None,padding:_size_2_t=0,ceil_mode:bool=False,count_include_pad:bool=True,divisor_override:Optional[int]=None)->None:super(AvgPool2d,self).__init__()self.kernel_size=kernel_sizeself.stride=strideif(strideisnotNone)elsekernel_sizeself.padding=paddingself.ceil_mode=ceil_modeself.count_include_pad=count_include_padself.divisor_override=divisor_overridedefforward(self,input:Tensor)->Tensor:returnF.avg_pool2d(input,self.kernel_size,self.stride,self.padding,self.ceil_mode,self.count_include_pad,self.divisor_override)
[docs]classAvgPool3d(_AvgPoolNd):r"""Applies a 3D average pooling over an input signal composed of several input planes. In the simplest case, the output value of the layer with input size :math:`(N, C, D, H, W)`, output :math:`(N, C, D_{out}, H_{out}, W_{out})` and :attr:`kernel_size` :math:`(kD, kH, kW)` can be precisely described as: .. math:: \begin{aligned} \text{out}(N_i, C_j, d, h, w) ={} & \sum_{k=0}^{kD-1} \sum_{m=0}^{kH-1} \sum_{n=0}^{kW-1} \\ & \frac{\text{input}(N_i, C_j, \text{stride}[0] \times d + k, \text{stride}[1] \times h + m, \text{stride}[2] \times w + n)} {kD \times kH \times kW} \end{aligned} If :attr:`padding` is non-zero, then the input is implicitly zero-padded on all three sides for :attr:`padding` number of points. Note: When ceil_mode=True, sliding windows are allowed to go off-bounds if they start within the left padding or the input. Sliding windows that would start in the right padded region are ignored. The parameters :attr:`kernel_size`, :attr:`stride` can either be: - a single ``int`` -- in which case the same value is used for the depth, height and width dimension - a ``tuple`` of three ints -- in which case, the first `int` is used for the depth dimension, the second `int` for the height dimension and the third `int` for the width dimension Args: kernel_size: the size of the window stride: the stride of the window. Default value is :attr:`kernel_size` padding: implicit zero padding to be added on all three sides ceil_mode: when True, will use `ceil` instead of `floor` to compute the output shape count_include_pad: when True, will include the zero-padding in the averaging calculation divisor_override: if specified, it will be used as divisor, otherwise :attr:`kernel_size` will be used Shape: - Input: :math:`(N, C, D_{in}, H_{in}, W_{in})` or :math:`(C, D_{in}, H_{in}, W_{in})`. - Output: :math:`(N, C, D_{out}, H_{out}, W_{out})` or :math:`(C, D_{out}, H_{out}, W_{out})`, where .. math:: D_{out} = \left\lfloor\frac{D_{in} + 2 \times \text{padding}[0] - \text{kernel\_size}[0]}{\text{stride}[0]} + 1\right\rfloor .. math:: H_{out} = \left\lfloor\frac{H_{in} + 2 \times \text{padding}[1] - \text{kernel\_size}[1]}{\text{stride}[1]} + 1\right\rfloor .. math:: W_{out} = \left\lfloor\frac{W_{in} + 2 \times \text{padding}[2] - \text{kernel\_size}[2]}{\text{stride}[2]} + 1\right\rfloor Examples:: >>> # pool of square window of size=3, stride=2 >>> m = nn.AvgPool3d(3, stride=2) >>> # pool of non-square window >>> m = nn.AvgPool3d((3, 2, 2), stride=(2, 1, 2)) >>> input = torch.randn(20, 16, 50,44, 31) >>> output = m(input) """__constants__=['kernel_size','stride','padding','ceil_mode','count_include_pad','divisor_override']kernel_size:_size_3_tstride:_size_3_tpadding:_size_3_tceil_mode:boolcount_include_pad:booldef__init__(self,kernel_size:_size_3_t,stride:Optional[_size_3_t]=None,padding:_size_3_t=0,ceil_mode:bool=False,count_include_pad:bool=True,divisor_override:Optional[int]=None)->None:super(AvgPool3d,self).__init__()self.kernel_size=kernel_sizeself.stride=strideif(strideisnotNone)elsekernel_sizeself.padding=paddingself.ceil_mode=ceil_modeself.count_include_pad=count_include_padself.divisor_override=divisor_overridedefforward(self,input:Tensor)->Tensor:returnF.avg_pool3d(input,self.kernel_size,self.stride,self.padding,self.ceil_mode,self.count_include_pad,self.divisor_override)def__setstate__(self,d):super(AvgPool3d,self).__setstate__(d)self.__dict__.setdefault('padding',0)self.__dict__.setdefault('ceil_mode',False)self.__dict__.setdefault('count_include_pad',True)
[docs]classFractionalMaxPool2d(Module):r"""Applies a 2D fractional max pooling over an input signal composed of several input planes. Fractional MaxPooling is described in detail in the paper `Fractional MaxPooling`_ by Ben Graham The max-pooling operation is applied in :math:`kH \times kW` regions by a stochastic step size determined by the target output size. The number of output features is equal to the number of input planes. Args: kernel_size: the size of the window to take a max over. Can be a single number k (for a square kernel of k x k) or a tuple `(kh, kw)` output_size: the target output size of the image of the form `oH x oW`. Can be a tuple `(oH, oW)` or a single number oH for a square image `oH x oH` output_ratio: If one wants to have an output size as a ratio of the input size, this option can be given. This has to be a number or tuple in the range (0, 1) return_indices: if ``True``, will return the indices along with the outputs. Useful to pass to :meth:`nn.MaxUnpool2d`. Default: ``False`` Shape: - Input: :math:`(N, C, H_{in}, W_{in})` or :math:`(C, H_{in}, W_{in})`. - Output: :math:`(N, C, H_{out}, W_{out})` or :math:`(C, H_{out}, W_{out})`, where :math:`(H_{out}, W_{out})=\text{output\_size}` or :math:`(H_{out}, W_{out})=\text{output\_ratio} \times (H_{in}, W_{in})`. Examples: >>> # pool of square window of size=3, and target output size 13x12 >>> m = nn.FractionalMaxPool2d(3, output_size=(13, 12)) >>> # pool of square window and target output size being half of input image size >>> m = nn.FractionalMaxPool2d(3, output_ratio=(0.5, 0.5)) >>> input = torch.randn(20, 16, 50, 32) >>> output = m(input) .. _Fractional MaxPooling: https://arxiv.org/abs/1412.6071 """__constants__=['kernel_size','return_indices','output_size','output_ratio']kernel_size:_size_2_treturn_indices:booloutput_size:_size_2_toutput_ratio:_ratio_2_tdef__init__(self,kernel_size:_size_2_t,output_size:Optional[_size_2_t]=None,output_ratio:Optional[_ratio_2_t]=None,return_indices:bool=False,_random_samples=None)->None:super(FractionalMaxPool2d,self).__init__()self.kernel_size=_pair(kernel_size)self.return_indices=return_indicesself.register_buffer('_random_samples',_random_samples)self.output_size=_pair(output_size)ifoutput_sizeisnotNoneelseNoneself.output_ratio=_pair(output_ratio)ifoutput_ratioisnotNoneelseNoneifoutput_sizeisNoneandoutput_ratioisNone:raiseValueError("FractionalMaxPool2d requires specifying either ""an output size, or a pooling ratio")ifoutput_sizeisnotNoneandoutput_ratioisnotNone:raiseValueError("only one of output_size and output_ratio may be specified")ifself.output_ratioisnotNone:ifnot(0<self.output_ratio[0]<1and0<self.output_ratio[1]<1):raiseValueError("output_ratio must be between 0 and 1 (got {})".format(output_ratio))defforward(self,input:Tensor):returnF.fractional_max_pool2d(input,self.kernel_size,self.output_size,self.output_ratio,self.return_indices,_random_samples=self._random_samples)
[docs]classFractionalMaxPool3d(Module):r"""Applies a 3D fractional max pooling over an input signal composed of several input planes. Fractional MaxPooling is described in detail in the paper `Fractional MaxPooling`_ by Ben Graham The max-pooling operation is applied in :math:`kT \times kH \times kW` regions by a stochastic step size determined by the target output size. The number of output features is equal to the number of input planes. Args: kernel_size: the size of the window to take a max over. Can be a single number k (for a square kernel of k x k x k) or a tuple `(kt x kh x kw)` output_size: the target output size of the image of the form `oT x oH x oW`. Can be a tuple `(oT, oH, oW)` or a single number oH for a square image `oH x oH x oH` output_ratio: If one wants to have an output size as a ratio of the input size, this option can be given. This has to be a number or tuple in the range (0, 1) return_indices: if ``True``, will return the indices along with the outputs. Useful to pass to :meth:`nn.MaxUnpool3d`. Default: ``False`` Shape: - Input: :math:`(N, C, T_{in}, H_{in}, W_{in})` or :math:`(C, T_{in}, H_{in}, W_{in})`. - Output: :math:`(N, C, T_{out}, H_{out}, W_{out})` or :math:`(C, T_{out}, H_{out}, W_{out})`, where :math:`(T_{out}, H_{out}, W_{out})=\text{output\_size}` or :math:`(T_{out}, H_{out}, W_{out})=\text{output\_ratio} \times (T_{in}, H_{in}, W_{in})` Examples: >>> # pool of cubic window of size=3, and target output size 13x12x11 >>> m = nn.FractionalMaxPool3d(3, output_size=(13, 12, 11)) >>> # pool of cubic window and target output size being half of input size >>> m = nn.FractionalMaxPool3d(3, output_ratio=(0.5, 0.5, 0.5)) >>> input = torch.randn(20, 16, 50, 32, 16) >>> output = m(input) .. _Fractional MaxPooling: https://arxiv.org/abs/1412.6071 """__constants__=['kernel_size','return_indices','output_size','output_ratio']kernel_size:_size_3_treturn_indices:booloutput_size:_size_3_toutput_ratio:_ratio_3_tdef__init__(self,kernel_size:_size_3_t,output_size:Optional[_size_3_t]=None,output_ratio:Optional[_ratio_3_t]=None,return_indices:bool=False,_random_samples=None)->None:super(FractionalMaxPool3d,self).__init__()self.kernel_size=_triple(kernel_size)self.return_indices=return_indicesself.register_buffer('_random_samples',_random_samples)self.output_size=_triple(output_size)ifoutput_sizeisnotNoneelseNoneself.output_ratio=_triple(output_ratio)ifoutput_ratioisnotNoneelseNoneifoutput_sizeisNoneandoutput_ratioisNone:raiseValueError("FractionalMaxPool3d requires specifying either ""an output size, or a pooling ratio")ifoutput_sizeisnotNoneandoutput_ratioisnotNone:raiseValueError("only one of output_size and output_ratio may be specified")ifself.output_ratioisnotNone:ifnot(0<self.output_ratio[0]<1and0<self.output_ratio[1]<1and0<self.output_ratio[2]<1):raiseValueError("output_ratio must be between 0 and 1 (got {})".format(output_ratio))defforward(self,input:Tensor):returnF.fractional_max_pool3d(input,self.kernel_size,self.output_size,self.output_ratio,self.return_indices,_random_samples=self._random_samples)
[docs]classLPPool1d(_LPPoolNd):r"""Applies a 1D power-average pooling over an input signal composed of several input planes. On each window, the function computed is: .. math:: f(X) = \sqrt[p]{\sum_{x \in X} x^{p}} - At p = :math:`\infty`, one gets Max Pooling - At p = 1, one gets Sum Pooling (which is proportional to Average Pooling) .. note:: If the sum to the power of `p` is zero, the gradient of this function is not defined. This implementation will set the gradient to zero in this case. Args: kernel_size: a single int, the size of the window stride: a single int, the stride of the window. Default value is :attr:`kernel_size` ceil_mode: when True, will use `ceil` instead of `floor` to compute the output shape Shape: - Input: :math:`(N, C, L_{in})` or :math:`(C, L_{in})`. - Output: :math:`(N, C, L_{out})` or :math:`(C, L_{out})`, where .. math:: L_{out} = \left\lfloor\frac{L_{in} - \text{kernel\_size}}{\text{stride}} + 1\right\rfloor Examples:: >>> # power-2 pool of window of length 3, with stride 2. >>> m = nn.LPPool1d(2, 3, stride=2) >>> input = torch.randn(20, 16, 50) >>> output = m(input) """kernel_size:_size_1_tstride:_size_1_tdefforward(self,input:Tensor)->Tensor:returnF.lp_pool1d(input,float(self.norm_type),self.kernel_size,self.stride,self.ceil_mode)
[docs]classLPPool2d(_LPPoolNd):r"""Applies a 2D power-average pooling over an input signal composed of several input planes. On each window, the function computed is: .. math:: f(X) = \sqrt[p]{\sum_{x \in X} x^{p}} - At p = :math:`\infty`, one gets Max Pooling - At p = 1, one gets Sum Pooling (which is proportional to average pooling) The parameters :attr:`kernel_size`, :attr:`stride` can either be: - a single ``int`` -- in which case the same value is used for the height and width dimension - a ``tuple`` of two ints -- in which case, the first `int` is used for the height dimension, and the second `int` for the width dimension .. note:: If the sum to the power of `p` is zero, the gradient of this function is not defined. This implementation will set the gradient to zero in this case. Args: kernel_size: the size of the window stride: the stride of the window. Default value is :attr:`kernel_size` ceil_mode: when True, will use `ceil` instead of `floor` to compute the output shape Shape: - Input: :math:`(N, C, H_{in}, W_{in})` - Output: :math:`(N, C, H_{out}, W_{out})`, where .. math:: H_{out} = \left\lfloor\frac{H_{in} - \text{kernel\_size}[0]}{\text{stride}[0]} + 1\right\rfloor .. math:: W_{out} = \left\lfloor\frac{W_{in} - \text{kernel\_size}[1]}{\text{stride}[1]} + 1\right\rfloor Examples:: >>> # power-2 pool of square window of size=3, stride=2 >>> m = nn.LPPool2d(2, 3, stride=2) >>> # pool of non-square window of power 1.2 >>> m = nn.LPPool2d(1.2, (3, 2), stride=(2, 1)) >>> input = torch.randn(20, 16, 50, 32) >>> output = m(input) """kernel_size:_size_2_tstride:_size_2_tdefforward(self,input:Tensor)->Tensor:returnF.lp_pool2d(input,float(self.norm_type),self.kernel_size,self.stride,self.ceil_mode)
class_AdaptiveMaxPoolNd(Module):__constants__=['output_size','return_indices']return_indices:booldef__init__(self,output_size:_size_any_opt_t,return_indices:bool=False)->None:super(_AdaptiveMaxPoolNd,self).__init__()self.output_size=output_sizeself.return_indices=return_indicesdefextra_repr(self)->str:return'output_size={}'.format(self.output_size)# FIXME (by @ssnl): Improve adaptive pooling docs: specify what the input and# output shapes are, and how the operation computes output.
[docs]classAdaptiveMaxPool1d(_AdaptiveMaxPoolNd):r"""Applies a 1D adaptive max pooling over an input signal composed of several input planes. The output size is :math:`L_{out}`, for any input size. The number of output features is equal to the number of input planes. Args: output_size: the target output size :math:`L_{out}`. return_indices: if ``True``, will return the indices along with the outputs. Useful to pass to nn.MaxUnpool1d. Default: ``False`` Shape: - Input: :math:`(N, C, L_{in})` or :math:`(C, L_{in})`. - Output: :math:`(N, C, L_{out})` or :math:`(C, L_{out})`, where :math:`L_{out}=\text{output\_size}`. Examples: >>> # target output size of 5 >>> m = nn.AdaptiveMaxPool1d(5) >>> input = torch.randn(1, 64, 8) >>> output = m(input) """output_size:_size_1_tdefforward(self,input:Tensor)->Tensor:returnF.adaptive_max_pool1d(input,self.output_size,self.return_indices)
[docs]classAdaptiveMaxPool2d(_AdaptiveMaxPoolNd):r"""Applies a 2D adaptive max pooling over an input signal composed of several input planes. The output is of size :math:`H_{out} \times W_{out}`, for any input size. The number of output features is equal to the number of input planes. Args: output_size: the target output size of the image of the form :math:`H_{out} \times W_{out}`. Can be a tuple :math:`(H_{out}, W_{out})` or a single :math:`H_{out}` for a square image :math:`H_{out} \times H_{out}`. :math:`H_{out}` and :math:`W_{out}` can be either a ``int``, or ``None`` which means the size will be the same as that of the input. return_indices: if ``True``, will return the indices along with the outputs. Useful to pass to nn.MaxUnpool2d. Default: ``False`` Shape: - Input: :math:`(N, C, H_{in}, W_{in})` or :math:`(C, H_{in}, W_{in})`. - Output: :math:`(N, C, H_{out}, W_{out})` or :math:`(C, H_{out}, W_{out})`, where :math:`(H_{out}, W_{out})=\text{output\_size}`. Examples: >>> # target output size of 5x7 >>> m = nn.AdaptiveMaxPool2d((5,7)) >>> input = torch.randn(1, 64, 8, 9) >>> output = m(input) >>> # target output size of 7x7 (square) >>> m = nn.AdaptiveMaxPool2d(7) >>> input = torch.randn(1, 64, 10, 9) >>> output = m(input) >>> # target output size of 10x7 >>> m = nn.AdaptiveMaxPool2d((None, 7)) >>> input = torch.randn(1, 64, 10, 9) >>> output = m(input) """output_size:_size_2_opt_tdefforward(self,input:Tensor):returnF.adaptive_max_pool2d(input,self.output_size,self.return_indices)
[docs]classAdaptiveMaxPool3d(_AdaptiveMaxPoolNd):r"""Applies a 3D adaptive max pooling over an input signal composed of several input planes. The output is of size :math:`D_{out} \times H_{out} \times W_{out}`, for any input size. The number of output features is equal to the number of input planes. Args: output_size: the target output size of the image of the form :math:`D_{out} \times H_{out} \times W_{out}`. Can be a tuple :math:`(D_{out}, H_{out}, W_{out})` or a single :math:`D_{out}` for a cube :math:`D_{out} \times D_{out} \times D_{out}`. :math:`D_{out}`, :math:`H_{out}` and :math:`W_{out}` can be either a ``int``, or ``None`` which means the size will be the same as that of the input. return_indices: if ``True``, will return the indices along with the outputs. Useful to pass to nn.MaxUnpool3d. Default: ``False`` Shape: - Input: :math:`(N, C, D_{in}, H_{in}, W_{in})` or :math:`(C, D_{in}, H_{in}, W_{in})`. - Output: :math:`(N, C, D_{out}, H_{out}, W_{out})` or :math:`(C, D_{out}, H_{out}, W_{out})`, where :math:`(D_{out}, H_{out}, W_{out})=\text{output\_size}`. Examples: >>> # target output size of 5x7x9 >>> m = nn.AdaptiveMaxPool3d((5,7,9)) >>> input = torch.randn(1, 64, 8, 9, 10) >>> output = m(input) >>> # target output size of 7x7x7 (cube) >>> m = nn.AdaptiveMaxPool3d(7) >>> input = torch.randn(1, 64, 10, 9, 8) >>> output = m(input) >>> # target output size of 7x9x8 >>> m = nn.AdaptiveMaxPool3d((7, None, None)) >>> input = torch.randn(1, 64, 10, 9, 8) >>> output = m(input) """output_size:_size_3_opt_tdefforward(self,input:Tensor):returnF.adaptive_max_pool3d(input,self.output_size,self.return_indices)
[docs]classAdaptiveAvgPool1d(_AdaptiveAvgPoolNd):r"""Applies a 1D adaptive average pooling over an input signal composed of several input planes. The output size is :math:`L_{out}`, for any input size. The number of output features is equal to the number of input planes. Args: output_size: the target output size :math:`L_{out}`. Shape: - Input: :math:`(N, C, L_{in})` or :math:`(C, L_{in})`. - Output: :math:`(N, C, L_{out})` or :math:`(C, L_{out})`, where :math:`L_{out}=\text{output\_size}`. Examples: >>> # target output size of 5 >>> m = nn.AdaptiveAvgPool1d(5) >>> input = torch.randn(1, 64, 8) >>> output = m(input) """output_size:_size_1_tdefforward(self,input:Tensor)->Tensor:returnF.adaptive_avg_pool1d(input,self.output_size)
[docs]classAdaptiveAvgPool2d(_AdaptiveAvgPoolNd):r"""Applies a 2D adaptive average pooling over an input signal composed of several input planes. The output is of size H x W, for any input size. The number of output features is equal to the number of input planes. Args: output_size: the target output size of the image of the form H x W. Can be a tuple (H, W) or a single H for a square image H x H. H and W can be either a ``int``, or ``None`` which means the size will be the same as that of the input. Shape: - Input: :math:`(N, C, H_{in}, W_{in})` or :math:`(C, H_{in}, W_{in})`. - Output: :math:`(N, C, S_{0}, S_{1})` or :math:`(C, S_{0}, S_{1})`, where :math:`S=\text{output\_size}`. Examples: >>> # target output size of 5x7 >>> m = nn.AdaptiveAvgPool2d((5,7)) >>> input = torch.randn(1, 64, 8, 9) >>> output = m(input) >>> # target output size of 7x7 (square) >>> m = nn.AdaptiveAvgPool2d(7) >>> input = torch.randn(1, 64, 10, 9) >>> output = m(input) >>> # target output size of 10x7 >>> m = nn.AdaptiveAvgPool2d((None, 7)) >>> input = torch.randn(1, 64, 10, 9) >>> output = m(input) """output_size:_size_2_opt_tdefforward(self,input:Tensor)->Tensor:returnF.adaptive_avg_pool2d(input,self.output_size)
[docs]classAdaptiveAvgPool3d(_AdaptiveAvgPoolNd):r"""Applies a 3D adaptive average pooling over an input signal composed of several input planes. The output is of size D x H x W, for any input size. The number of output features is equal to the number of input planes. Args: output_size: the target output size of the form D x H x W. Can be a tuple (D, H, W) or a single number D for a cube D x D x D. D, H and W can be either a ``int``, or ``None`` which means the size will be the same as that of the input. Shape: - Input: :math:`(N, C, D_{in}, H_{in}, W_{in})` or :math:`(C, D_{in}, H_{in}, W_{in})`. - Output: :math:`(N, C, S_{0}, S_{1}, S_{2})` or :math:`(C, S_{0}, S_{1}, S_{2})`, where :math:`S=\text{output\_size}`. Examples: >>> # target output size of 5x7x9 >>> m = nn.AdaptiveAvgPool3d((5,7,9)) >>> input = torch.randn(1, 64, 8, 9, 10) >>> output = m(input) >>> # target output size of 7x7x7 (cube) >>> m = nn.AdaptiveAvgPool3d(7) >>> input = torch.randn(1, 64, 10, 9, 8) >>> output = m(input) >>> # target output size of 7x9x8 >>> m = nn.AdaptiveAvgPool3d((7, None, None)) >>> input = torch.randn(1, 64, 10, 9, 8) >>> output = m(input) """output_size:_size_3_opt_tdefforward(self,input:Tensor)->Tensor:returnF.adaptive_avg_pool3d(input,self.output_size)
Docs
Access comprehensive developer documentation for PyTorch
To analyze traffic and optimize your experience, we serve cookies on this site. By clicking or navigating, you agree to allow our usage of cookies. As the current maintainers of this site, Facebook’s Cookies Policy applies. Learn more, including about available controls: Cookies Policy.
