pytorch Architecture Docs
Home / Class/ TransposeConvTensorUnpackConversion Class — pytorch Architecture

TransposeConvTensorUnpackConversion Class — pytorch Architecture

Architecture documentation for the TransposeConvTensorUnpackConversion class in fbgemm_utils.cpp from the pytorch codebase.

Class cpp

Entity Profile

Source Code

aten/src/ATen/native/quantized/cpu/fbgemm_utils.cpp lines 247–254

template <>
Tensor TransposeConvTensorUnpackConversion<2>(const Tensor& src, int groups) {
  // OC IC/G HW -> IC OC/G HW logically
  auto oc_g_ic_g_hw_tensors = src.chunk(groups);
  auto fused_tensor = at::cat(oc_g_ic_g_hw_tensors, 1);
  set_quantizer_(fused_tensor, src.quantizer());
  return fused_tensor.permute({1, 0, 2, 3});
}

Source

Analyze Your Own Codebase

Get architecture documentation, dependency graphs, and domain analysis for your codebase in minutes.

Try Supermodel Free