Skip to content

Remove usage of deprecated get_ie_output_name() function #933

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
wants to merge 5 commits into
base: master
Choose a base branch
from
+1 −14
Open

Remove usage of deprecated get_ie_output_name() function #933

Show file tree
Hide file tree
Changes from 3 commits
Commits
Show all changes
5 commits
Select commit Hold shift + click to select a range
5db9413
Remove usage of deprecated get_ie_output_name() function
CuriousPanCake Sep 30, 2024
569a0cb
fix OPENVINO_DEBUG
CuriousPanCake Oct 1, 2024
85daf65
fix ;
CuriousPanCake Oct 1, 2024
fbb6fe9
Merge branch 'master' into legacy_fix
CuriousPanCake Feb 13, 2025
18f170a
Merge branch 'master' into legacy_fix
andrei-kochin May 12, 2025
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
6 changes: 3 additions & 3 deletions modules/llama_cpp_plugin/src/compiled_model.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -27,11 +27,11 @@ LlamaCppModel::LlamaCppModel(const std::string& gguf_fname,
: ICompiledModel(nullptr, plugin),
m_gguf_fname(gguf_fname),
m_num_threads(num_threads) {
OPENVINO_DEBUG << "llama_cpp_plugin: loading llama model directly from GGUF... " << std::endl;
OPENVINO_DEBUG("llama_cpp_plugin: loading llama model directly from GGUF... ");
llama_model_params mparams = llama_model_default_params();
mparams.n_gpu_layers = 99;
m_llama_model_ptr = llama_load_model_from_file(gguf_fname.c_str(), mparams);
OPENVINO_DEBUG << "llama_cpp_plugin: llama model loaded successfully from GGUF..." << std::endl;
OPENVINO_DEBUG("llama_cpp_plugin: llama model loaded successfully from GGUF...");

auto input_ids = std::make_shared<ov::opset13::Parameter>(ov::element::Type_t::i64, ov::PartialShape({-1, -1}));
auto fake_convert = std::make_shared<ov::opset13::Convert>(input_ids->output(0), ov::element::Type_t::f32);
Expand Down Expand Up @@ -71,7 +71,7 @@ std::shared_ptr<const ov::Model> LlamaCppModel::get_runtime_model() const {
}

void LlamaCppModel::set_property(const ov::AnyMap& properties) {
OPENVINO_DEBUG << "llama_cpp_plugin: attempted to set_property (did nothing)";
OPENVINO_DEBUG("llama_cpp_plugin: attempted to set_property (did nothing)");
}

ov::Any LlamaCppModel::get_property(const std::string& name) const {
Expand Down
8 changes: 4 additions & 4 deletions modules/llama_cpp_plugin/src/infer_request.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -28,7 +28,7 @@ void allocate_tensor_impl(ov::SoPtr<ov::ITensor>& tensor,
LlamaCppSyncInferRequest::LlamaCppSyncInferRequest(const std::shared_ptr<const LlamaCppModel>& compiled_model,
size_t num_threads)
: ov::ISyncInferRequest(compiled_model) {
OPENVINO_DEBUG << "llama_cpp_plugin: infer request ctor called\n";
OPENVINO_DEBUG("llama_cpp_plugin: infer request ctor called");
llama_context_params cparams = llama_context_default_params();
cparams.n_threads = num_threads ? num_threads : std::thread::hardware_concurrency();
cparams.n_ctx = 0; // this means that the actual n_ctx will be taken equal to the model's train-time value
Expand All @@ -51,7 +51,7 @@ LlamaCppSyncInferRequest::LlamaCppSyncInferRequest(const std::shared_ptr<const L
}
void LlamaCppSyncInferRequest::set_tensors_impl(const ov::Output<const ov::Node> port,
const std::vector<ov::SoPtr<ov::ITensor>>& tensors) {
OPENVINO_DEBUG << "llama_cpp_plugin: set_tensors_impl called\n";
OPENVINO_DEBUG("llama_cpp_plugin: set_tensors_impl called");
}

void llama_batch_add_reimpl(struct llama_batch& batch,
Expand Down Expand Up @@ -131,12 +131,12 @@ void LlamaCppSyncInferRequest::infer() {
llama_batch_free(batch);
};
std::vector<ov::ProfilingInfo> LlamaCppSyncInferRequest::get_profiling_info() const {
OPENVINO_DEBUG << "llama_cpp_plugin: get_profiling_info() called\n";
OPENVINO_DEBUG("llama_cpp_plugin: get_profiling_info() called");
return std::vector<ov::ProfilingInfo>{};
};

std::vector<ov::SoPtr<ov::IVariableState>> LlamaCppSyncInferRequest::query_state() const {
OPENVINO_DEBUG << "llama_cpp_plugin: query_state() called\n";
OPENVINO_DEBUG("llama_cpp_plugin: query_state() called");
return {std::static_pointer_cast<ov::IVariableState>(std::make_shared<LlamaCppState>(m_llama_ctx))};
}

Expand Down
15 changes: 1 addition & 14 deletions modules/nvidia_plugin/src/ops/result.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -62,21 +62,8 @@ std::optional<std::size_t> ResultOp::GetOutputTensorSubIndex(const ov::Output<ov
}

std::vector<std::string> ResultOp::GetOutputTensorName(const ov::op::v0::Result& node) {
std::vector<std::string> outputNames;

const auto& input = node.input_value(0);
auto name = ov::op::util::get_ie_output_name(input);
outputNames.push_back(name);

auto resultName = node.get_friendly_name();

// NOTE: New way of getting the fused names for OpenVINO 2.0 API
// TODO: When support for old OpenVINO API will be stopped, consider using only this approach.
// Also see any issues with Tacatron2 network
const auto& fusedResults = ov::getFusedNamesVector(input.get_node()->shared_from_this());
outputNames.insert(outputNames.end(), fusedResults.begin(), fusedResults.end());

return outputNames;
return ov::getFusedNamesVector(input.get_node()->shared_from_this());
}

void ResultOp::Capture(InferenceRequestContext& context,
Expand Down
Loading