MakeConvOutputShape Class — pytorch Architecture
Architecture documentation for the MakeConvOutputShape class in ConvUtils.h from the pytorch codebase.
Entity Profile
Source Code
aten/src/ATen/native/quantized/ConvUtils.h lines 21–37
template <>
at::SmallVector<int64_t, 4> MakeConvOutputShape<2>(
int N, // mini-batch
int M, // output channels
const std::array<int64_t, 2>& input_image_shape,
const std::vector<int64_t>& kernel,
const at::List<int64_t>& stride,
const at::List<int64_t>& padding,
const at::List<int64_t>& dilation) {
const int H = input_image_shape[0];
const int W = input_image_shape[1];
const int64_t Y_H =
(H + 2 * padding[0] - dilation[0] * (kernel[0] - 1) - 1) / stride[0] + 1;
const int64_t Y_W =
(W + 2 * padding[1] - dilation[1] * (kernel[1] - 1) - 1) / stride[1] + 1;
return {N, M, Y_H, Y_W};
}
Source
Analyze Your Own Codebase
Get architecture documentation, dependency graphs, and domain analysis for your codebase in minutes.
Try Supermodel Free