Deprecated List

Class InferenceEngine::BatchNormalizationLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::BatchToSpaceLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::BinaryConvolutionLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Global InferenceEngine::Blob::element_size () const =0

Cast to MemoryBlob and use its API instead. Blob class can represent compound blob, which do not refer to the only solid memory.

Global InferenceEngine::Blob::product (const SizeVector &dims) noexcept

Cast to MemoryBlob and use its API instead.

Global InferenceEngine::Blob::properProduct (const SizeVector &dims) noexcept

Cast to MemoryBlob and use its API instead.

Class InferenceEngine::BroadcastLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::BucketizeLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ClampLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::CNNLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Global InferenceEngine::CNNNetwork::CNNNetwork (std::shared_ptr< ICNNNetwork > network)

Don’t use this constructor. It will be removed soon

Global InferenceEngine::CNNNetwork::operator const ICNNNetwork & () const

InferenceEngine::ICNNNetwork interface is deprecated

Global InferenceEngine::CNNNetwork::operator ICNNNetwork & ()

InferenceEngine::ICNNNetwork interface is deprecated

Global InferenceEngine::CNNNetwork::operator ICNNNetwork::Ptr ()

InferenceEngine::ICNNNetwork interface is deprecated

Class InferenceEngine::ConcatLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ConvolutionLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Global InferenceEngine::Core::ImportNetwork (std::istream &networkModel)

Use Core::ImportNetwork with explicit device name

Class InferenceEngine::CropLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Global InferenceEngine::Data::reshape (const std::initializer_list< size_t > &dims, Layout layout)

Use InferenceEngine::Data::reshape(const SizeVector&, Layout)

Class InferenceEngine::DeconvolutionLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::DeformableConvolutionLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::DepthToSpaceLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::EltwiseLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Global InferenceEngine::ExecutableNetwork::CreateInferRequestPtr ()

Use ExecutableNetwork::CreateInferRequest

Global InferenceEngine::ExecutableNetwork::operator std::shared_ptr< IExecutableNetwork > ()

Will be removed. Use operator bool

Global InferenceEngine::ExecutableNetwork::reset (std::shared_ptr< IExecutableNetwork > newActual)

The method Will be removed

Class InferenceEngine::ExperimentalDetectronPriorGridGeneratorLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ExperimentalSparseWeightedReduceLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::FillLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::FullyConnectedLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::GatherLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::GemmLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::GRNLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::GRUCell

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ICNNNetwork

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::addOutput (const std::string &layerName, size_t outputIndex=0, ResponseDesc *resp=nullptr) noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::getBatchSize () const =0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::getFunction () const noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::getFunction () noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::getInput (const std::string &inputName) const noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::getInputsInfo (InputsDataMap &inputs) const noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::getName () const noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::getOutputsInfo (OutputsDataMap &out) const noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::getOVNameForTensor (std::string &ov_name, const std::string &orig_name, ResponseDesc *resp) const noexcept

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::InputShapes

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::layerCount () const =0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::Ptr

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::reshape (const std::map< std::string, ngraph::PartialShape > &partialShapes, ResponseDesc *resp) noexcept

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::reshape (const InputShapes &inputShapes, ResponseDesc *resp) noexcept

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::serialize (std::ostream &xmlStream, Blob::Ptr &binData, ResponseDesc *resp) const noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::serialize (std::ostream &xmlStream, std::ostream &binStream, ResponseDesc *resp) const noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::serialize (const std::string &xmlPath, const std::string &binPath, ResponseDesc *resp) const noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::ICNNNetwork::setBatchSize (size_t size, ResponseDesc *responseDesc) noexcept=0

Use InferenceEngine::CNNNetwork wrapper instead

Global InferenceEngine::IExecutableNetwork::GetExecGraphInfo (ICNNNetwork::Ptr &graphPtr, ResponseDesc *resp) noexcept=0

Use InferenceEngine::ExecutableNetwork::GetExecGraphInfo instead

Global InferenceEngine::IExecutableNetworkInternal::Export (const std::string &modelFileName)

Use IExecutableNetworkInternal::Export(std::ostream& networkModel)

Global InferenceEngine::IInferencePlugin::ImportNetwork (const std::string &modelFileName, const std::map< std::string, std::string > &config)

Use ImportNetwork(std::istream& networkModel, const std::map<std::string, std::string>& config)

Class InferenceEngine::IInferRequest

Use InferenceEngine::InferRequest C++ wrapper

Class InferenceEngine::LayerParams

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Global InferenceEngine::LowLatency (InferenceEngine::CNNNetwork &network)

Use InferenceEngine::lowLatency2 instead. This transformation will be removed in 2023.1.

Class InferenceEngine::LSTMCell

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::MathLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::MVNLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::NonMaxSuppressionLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::NormLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::OneHotLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::PadLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Global InferenceEngine::PluginConfigParams::KEY_DUMP_EXEC_GRAPH_AS_DOT

Use InferenceEngine::ExecutableNetwork::GetExecGraphInfo::serialize method

Class InferenceEngine::PoolingLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::PowerLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::PReLULayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::QuantizeLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::RangeLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ReduceLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ReLU6Layer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ReLULayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ReshapeLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ReverseSequenceLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::RNNCell

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::RNNCellBase

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::RNNSequenceLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ScaleShiftLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ScatterElementsUpdateLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ScatterUpdateLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::SelectLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::ShuffleChannelsLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::SoftMaxLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::SpaceToBatchLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::SpaceToDepthLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::SparseFillEmptyRowsLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::SparseSegmentReduceLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::SparseToDenseLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::SplitLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::StridedSliceLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::TensorIterator

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::TileLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::TopKLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::UniqueLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class InferenceEngine::Version::ApiVersion

Use IE_VERSION_[MAJOR|MINOR|PATCH] definitions, buildNumber property

Class InferenceEngine::WeightableLayer

Migrate to IR v10 and work with ngraph::Function directly. The method will be removed in 2021.1

Class ngraph::CoordinateIterator

Global ngraph::CoordinateTransformBasic::index (const Coordinate &c) const

Global ngraph::maximum_value (const Output< Node > &value)

Use evaluate_upper_bound instead

Global ov::Core::add_extension (const std::shared_ptr< InferenceEngine::IExtension > &extension)

This method is deprecated. Please use other Core::add_extension methods.

Global ov::Model::evaluate (const ov::HostTensorVector &output_tensors, const ov::HostTensorVector &input_tensors) const

Use evaluate with ov::Tensor instead

Global ov::Model::evaluate (const ov::HostTensorVector &output_tensors, const ov::HostTensorVector &input_tensors, ov::EvaluationContext &evaluation_context) const

Use evaluate with ov::Tensor instead

Global ov::Node::evaluate (const ov::HostTensorVector &output_values, const ov::HostTensorVector &input_values, const EvaluationContext &evaluationContext) const

Use evaluate with ov::Tensor instead

Global ov::Node::evaluate (const ov::HostTensorVector &output_values, const ov::HostTensorVector &input_values) const

Use evaluate with ov::Tensor instead