MaxPool1d#

classtorch.nn.modules.pooling.MaxPool1d(kernel_size,stride=None,padding=0,dilation=1,return_indices=False,ceil_mode=False)[source]#

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$(N,C,L)$(N,C,L)and output$(N,C,L_{out})$(N,C,Lout​) can be precisely described as:

$$out(N_i,C_j,k)=\underset{m=0,\dots ,\text{kernel\_size}-1}{\max }input(N_i,C_j,stride\times k+m)$$out(Ni​,Cj​,k)=m=0,…,kernel_size−1max​input(Ni​,Cj​,stride×k+m)

Ifpadding is non-zero, then the input is implicitly padded with negative infinity on both sidesforpadding number of points.dilation is the stride between the elements within thesliding window. Thislink 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 paddingor the input. Sliding windows that would start in the right padded region are ignored.

Parameters

• kernel_size (Union[int,tuple[int]]) – The size of the sliding window, must be > 0.

• stride (Union[int,tuple[int]]) – The stride of the sliding window, must be > 0. Default value iskernel_size.

• padding (Union[int,tuple[int]]) – Implicit negative infinity padding to be added on both sides, must be >= 0 and <= kernel_size / 2.

• dilation (Union[int,tuple[int]]) – The stride between elements within a sliding window, must be > 0.

• return_indices (bool) – IfTrue, will return the argmax along with the max values.Useful fortorch.nn.MaxUnpool1d later

• ceil_mode (bool) – IfTrue, will useceil instead offloor to compute the output shape. Thisensures that every element in the input tensor is covered by a sliding window.

Shape:

• Input:$(N,C,L_{in})$(N,C,Lin​) or$(C,L_{in})$(C,Lin​).

• Output:$(N,C,L_{out})$(N,C,Lout​) or$(C,L_{out})$(C,Lout​),

  whereceil_mode=False

  $$L_{out}=\lfloor \frac{L_{in}+2\times \text{padding}-\text{dilation}\times (\text{kernel\_size}-1)-1}{\text{stride}}\rfloor +1$$Lout​=⌊strideLin​+2×padding−dilation×(kernel_size−1)−1​⌋+1

  whereceil_mode=True

  $$L_{out}=\lceil \frac{L_{in}+2\times \text{padding}-\text{dilation}\times (\text{kernel\_size}-1)-1+(stride-1)}{\text{stride}}\rceil +1$$Lout​=⌈strideLin​+2×padding−dilation×(kernel_size−1)−1+(stride−1)​⌉+1

• Ensure that the last pooling starts inside the image, make$L_{out}=L_{out}-1$Lout​=Lout​−1when$(L_{out}-1)\ast \text{stride}>=L_{in}+\text{padding}$(Lout​−1)∗stride>=Lin​+padding.

Examples:

>>># pool of size=3, stride=2>>>m=nn.MaxPool1d(3,stride=2)>>>input=torch.randn(20,16,50)>>>output=m(input)

forward(input)[source]#

Runs the forward pass.