update_from Class — pytorch Architecture
Architecture documentation for the update_from class in DistributionTemplates.h from the pytorch codebase.
Entity Profile
Source Code
aten/src/ATen/native/DistributionTemplates.h lines 42–57
template<typename scalar_t>
int64_t update_from(int64_t from) {
static_assert(
std::is_floating_point_v<scalar_t> ||
std::is_same_v<scalar_t, at::Half> ||
std::is_same_v<scalar_t, at::BFloat16>, "scalar_t must be floating-point type");
const auto from_plus_1 = static_cast<int64_t>(static_cast<scalar_t>(from + 1));
if (from_plus_1 < from) {
int64_t from_ = std::abs(from + 1);
int n = 0;
while (from_ >>= 1) ++n;
// NOLINTNEXTLINE(clang-analyzer-core.UndefinedBinaryOperatorResult)
from = from_plus_1 + (1LL << (n - std::numeric_limits<scalar_t>::digits + 1));
}
return from;
}
Source
Analyze Your Own Codebase
Get architecture documentation, dependency graphs, and domain analysis for your codebase in minutes.
Try Supermodel Free