Where can I find the source code of "max_pool2d"

#1

Hi there,

I’d like to find the source code of “max_pool2d” and find that actually Pytorch uses aten::max_pool2d_with_indices() to perform the operation, which is under the directory of aten/src/ATen/native/Pooling.cpp. See the code below:

std::tuple<Tensor, Tensor> max_pool1d_with_indices(
    const Tensor& self,
    IntArrayRef kernel_size,
    IntArrayRef stride,
    IntArrayRef padding,
    IntArrayRef dilation,
    bool ceil_mode) {
  if (stride.empty()) {
    stride = kernel_size;
  }
  checkDim("max_pool1d", TensorArg(self, "self", 1), 3);
  check1d("max_pool1d", "kernel_size", kernel_size);
  check1d("max_pool1d", "stride", stride);
  check1d("max_pool1d", "padding", padding);
  check1d("max_pool1d", "dilation", dilation);

  Tensor output, indices;
  std::tie(output, indices) = at::max_pool2d_with_indices(
      self.unsqueeze(2),
      {1, kernel_size[0]},
      {1, stride[0]},
      {0, padding[0]},
      {1, dilation[0]},
      ceil_mode);

  return std::make_tuple(output.squeeze(2), indices.squeeze(2));
}

However, it calls at::max_pool2d_with_indices() and where can I find this function specifically?
I know Pytorch uses Aten library for math computation but it seems that max_pool2d_with_indices() is the custom function using some basic maths.

Thanks!