apply_op Class — pytorch Architecture
Architecture documentation for the apply_op class in CPUApplyUtils.h from the pytorch codebase.
Entity Profile
Source Code
aten/src/ATen/CPUApplyUtils.h lines 234–256
template <typename Op, typename... Args>
inline void apply_op(
int64_t numel,
int64_t offset,
const Op& op,
Args... iters) {
// For 0-dim tensors
if (numel == 1 && max_dim(iters...) == 0) {
op(*iters.data_...);
return;
}
if (offset > 0)
forward(offset, iters...);
// Splitting this into chunks helps the compiler create faster assembly
for (int64_t i = 0; i < numel;) {
for (; iterate_continue(iters...) && i < numel;) {
op(*iters.data_...);
iterate(1, iters...);
i++;
}
iterate_overflow(iters...);
}
}
Source
Analyze Your Own Codebase
Get architecture documentation, dependency graphs, and domain analysis for your codebase in minutes.
Try Supermodel Free