Skip to content

Use a struct to manage the mapping between AttentionImpl enum values and their string names, simplifying GetAttentionImplName function. Add a test to ensure all valid AttentionImpl enums have a corresponding name and can be looked up. #844

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
copybara-service wants to merge 1 commit into
base: dev
Choose a base branch
from
Open

Use a struct to manage the mapping between AttentionImpl enum values and their string names, simplifying GetAttentionImplName function. Add a test to ensure all valid AttentionImpl enums have a corresponding name and can be looked up. #844

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
31 changes: 19 additions & 12 deletions gemma/configs.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -19,6 +19,7 @@
#include <stdio.h>

#include <string>
#include <utility>
#include <vector>

#include "compression/types.h" // Type
Expand Down Expand Up @@ -678,7 +679,7 @@ Model DeduceModel(const Path& blob_path, size_t layers, int layer_types) {
return Model::GEMMA3_270M;

case 26:
if (layer_types & (kDeducedViT|kDeducedKqNorm)) {
if (layer_types & (kDeducedViT | kDeducedKqNorm)) {
return Model::GEMMA3_1B;
}
return Model::GEMMA2_2B;
Expand Down Expand Up @@ -712,22 +713,28 @@ Model DeduceModel(const Path& blob_path, size_t layers, int layer_types) {
}
}

// Keep in sync with enum class AttentionImpl.
const char* kAttentionImplNames[] = {
"old", "flash",
"unknown" // keep last
constexpr std::pair<const char*, AttentionImpl> kAttentionImplNameToEnum[] = {
{"old", AttentionImpl::kOld},
{"flash", AttentionImpl::kFlash},
{"flash_transposed_qs", AttentionImpl::kFlashTransposedQs},
{"flash_transposed_qs_bf16", AttentionImpl::kFlashTransposedQsBF16},
};

std::string GetAttentionImplName(AttentionImpl impl) {
return kAttentionImplNames[static_cast<size_t>(impl)];
for (auto const& [name, attention_impl] : kAttentionImplNameToEnum) {
if (attention_impl == impl) return std::string(name);
}
return "unknown";
}

AttentionImpl GetAttentionImpl(const std::string& impl) {
if (impl == GetAttentionImplName(AttentionImpl::kOld))
return AttentionImpl::kOld;
if (impl == GetAttentionImplName(AttentionImpl::kFlash))
return AttentionImpl::kFlash;
HWY_WARN("Unknown attention implementation: %s. Using kOld.\n", impl.c_str());
AttentionImpl GetAttentionImpl(const std::string& impl_name) {
for (auto const& [name, attention_impl] : kAttentionImplNameToEnum) {
if (name == impl_name) {
return attention_impl;
}
}
HWY_WARN("Unknown attention implementation: %s. Using kOld.\n",
impl_name.c_str());
return AttentionImpl::kOld;
}

Expand Down
12 changes: 12 additions & 0 deletions gemma/configs_test.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -41,4 +41,16 @@ TEST(ConfigsTest, TestAll) {
});
}

TEST(ConfigsTest, TestAttentionImpl) {
for (int i = 0; i < static_cast<int>(AttentionImpl::kSentinel); ++i) {
AttentionImpl impl = static_cast<AttentionImpl>(i);
std::string name = GetAttentionImplName(impl);
ASSERT_NE(name, "unknown");
ASSERT_EQ(GetAttentionImpl(name), impl);
}
ASSERT_EQ(GetAttentionImplName(AttentionImpl::kSentinel), "unknown");
ASSERT_EQ(GetAttentionImpl("unknown"), AttentionImpl::kOld);
ASSERT_EQ(GetAttentionImpl("invalid"), AttentionImpl::kOld);
}

} // namespace gcpp
Loading