Class Net

java.lang.Object
org.opencv.dnn.Net
public class Net extends Object
This class allows to create and manipulate comprehensive artificial neural networks. Neural network is presented as directed acyclic graph (DAG), where vertices are Layer instances, and edges specify relationships between layers inputs and outputs. Each network layer has unique integer id and unique string name inside its network. LayerId can store either layer name or layer id. This class supports reference counting of its instances, i. e. copies point to the same instance.

  • Field Details

    • nativeObj

      protected final long nativeObj

  • Constructor Details

    • Net

      protected Net(long addr)

    • Net

      public Net()

  • Method Details

    • getNativeObjAddr

      public long getNativeObjAddr()

    • __fromPtr__

      public static Net __fromPtr__(long addr)

    • readFromModelOptimizer

      public static Net readFromModelOptimizer(String xml, String bin)
      Create a network from Intel's Model Optimizer intermediate representation (IR).
      Parameters:
      xml - XML configuration file with network's topology.
      bin - Binary file with trained weights. Networks imported from Intel's Model Optimizer are launched in Intel's Inference Engine backend.
      Returns:
      automatically generated

    • readFromModelOptimizer

      public static Net readFromModelOptimizer(MatOfByte bufferModelConfig, MatOfByte bufferWeights)
      Create a network from Intel's Model Optimizer in-memory buffers with intermediate representation (IR).
      Parameters:
      bufferModelConfig - buffer with model's configuration.
      bufferWeights - buffer with model's trained weights.
      Returns:
      Net object.

    • empty

      public boolean empty()
      Returns true if there are no layers in the network.
      Returns:
      automatically generated

    • dump

      public String dump()
      Dump net to String
      Returns:
      String with structure, hyperparameters, backend, target and fusion Call method after setInput(). To see correct backend, target and fusion run after forward().

    • dumpToFile

      public void dumpToFile(String path)
      Dump net structure, hyperparameters, backend, target and fusion to dot file
      Parameters:
      path - path to output file with .dot extension SEE: dump()

    • dumpToPbtxt

      public void dumpToPbtxt(String path)
      Dump net structure, hyperparameters, backend, target and fusion to pbtxt file
      Parameters:
      path - path to output file with .pbtxt extension Use Netron (https://netron.app) to open the target file to visualize the model. Call method after setInput(). To see correct backend, target and fusion run after forward().

    • getLayerId

      public int getLayerId(String layer)
      Converts string name of the layer to the integer identifier.
      Parameters:
      layer - automatically generated
      Returns:
      id of the layer, or -1 if the layer wasn't found.

    • getLayerNames

      public List<String> getLayerNames()

    • getLayer

      public Layer getLayer(int layerId)
      Returns pointer to layer with specified id or name which the network use.
      Parameters:
      layerId - automatically generated
      Returns:
      automatically generated

    • getLayer

      @Deprecated public Layer getLayer(String layerName)
      Deprecated.
      Use int getLayerId(const String &layer)
      Parameters:
      layerName - automatically generated
      Returns:
      automatically generated

    • getLayer

      @Deprecated public Layer getLayer(DictValue layerId)
      Deprecated.
      to be removed
      Parameters:
      layerId - automatically generated
      Returns:
      automatically generated

    • connect

      public void connect(String outPin, String inpPin)
      Connects output of the first layer to input of the second layer.
      Parameters:
      outPin - descriptor of the first layer output.
      inpPin - descriptor of the second layer input. Descriptors have the following template <DFN>&lt;layer_name&gt;[.input_number]</DFN>: - the first part of the template <DFN>layer_name</DFN> is string name of the added layer. If this part is empty then the network input pseudo layer will be used; - the second optional part of the template <DFN>input_number</DFN> is either number of the layer input, either label one. If this part is omitted then the first layer input will be used. SEE: setNetInputs(), Layer::inputNameToIndex(), Layer::outputNameToIndex()

    • registerOutput

      public int registerOutput(String outputName, int layerId, int outputPort)
      Registers network output with name Function may create additional 'Identity' layer.
      Parameters:
      outputName - identifier of the output
      layerId - identifier of the second layer
      outputPort - number of the second layer input
      Returns:
      index of bound layer (the same as layerId or newly created)

    • setInputsNames

      public void setInputsNames(List<String> inputBlobNames)
      Sets outputs names of the network input pseudo layer. Each net always has special own the network input pseudo layer with id=0. This layer stores the user blobs only and don't make any computations. In fact, this layer provides the only way to pass user data into the network. As any other layer, this layer can label its outputs and this function provides an easy way to do this.
      Parameters:
      inputBlobNames - automatically generated

    • setInputShape

      public void setInputShape(String inputName, MatOfInt shape)
      Specify shape of network input.
      Parameters:
      inputName - automatically generated
      shape - automatically generated

    • forward

      public Mat forward(String outputName)
      Runs forward pass to compute output of layer with name outputName.
      Parameters:
      outputName - name for layer which output is needed to get
      Returns:
      blob for first output of specified layer. By default runs forward pass for the whole network.

    • forward

      public Mat forward()
      Runs forward pass to compute output of layer with name outputName.
      Returns:
      blob for first output of specified layer. By default runs forward pass for the whole network.

    • forward

      public void forward(List<Mat> outputBlobs, String outputName)
      Runs forward pass to compute output of layer with name outputName.
      Parameters:
      outputBlobs - contains all output blobs for specified layer.
      outputName - name for layer which output is needed to get If outputName is empty, runs forward pass for the whole network.

    • forward

      public void forward(List<Mat> outputBlobs)
      Runs forward pass to compute output of layer with name outputName.
      Parameters:
      outputBlobs - contains all output blobs for specified layer. If outputName is empty, runs forward pass for the whole network.

    • forward

      public void forward(List<Mat> outputBlobs, List<String> outBlobNames)
      Runs forward pass to compute outputs of layers listed in outBlobNames.
      Parameters:
      outputBlobs - contains blobs for first outputs of specified layers.
      outBlobNames - names for layers which outputs are needed to get

    • forwardAndRetrieve

      public void forwardAndRetrieve(List<List<Mat>> outputBlobs, List<String> outBlobNames)
      Runs forward pass to compute outputs of layers listed in outBlobNames.
      Parameters:
      outputBlobs - contains all output blobs for each layer specified in outBlobNames.
      outBlobNames - names for layers which outputs are needed to get

    • setPreferableBackend

      public void setPreferableBackend(int backendId)
      Ask network to use specific computation backend where it supported.
      Parameters:
      backendId - backend identifier. SEE: Backend

    • setPreferableTarget

      public void setPreferableTarget(int targetId)
      Ask network to make computations on specific target device.
      Parameters:
      targetId - target identifier. SEE: Target List of supported combinations backend / target: | | DNN_BACKEND_OPENCV | DNN_BACKEND_INFERENCE_ENGINE | DNN_BACKEND_CUDA | |------------------------|--------------------|------------------------------|-------------------| | DNN_TARGET_CPU | + | + | | | DNN_TARGET_OPENCL | + | + | | | DNN_TARGET_OPENCL_FP16 | + | + | | | DNN_TARGET_MYRIAD | | + | | | DNN_TARGET_FPGA | | + | | | DNN_TARGET_CUDA | | | + | | DNN_TARGET_CUDA_FP16 | | | + | | DNN_TARGET_HDDL | | + | |

    • finalizeNet

      public void finalizeNet()
      Finalizes the network configuration and prepares it for inference. This method must be called after setting backend/target via setPreferableBackend() and setPreferableTarget(), and before the first forward() call. It creates the underlying execution session (e.g. ONNX Runtime session) on the configured backend/target. If not called explicitly, the first forward() will call it automatically. Calling finalizeNet() early lets you pay the one-time setup cost at a predictable point and catch configuration errors before inference.

    • setInput

      public void setInput(Mat blob, String name, double scalefactor, Scalar mean)
      Sets the new input value for the network
      Parameters:
      blob - A new blob. Should have CV_32F or CV_8U depth.
      name - A name of input layer.
      scalefactor - An optional normalization scale.
      mean - An optional mean subtraction values. SEE: connect(String, String) to know format of the descriptor. If scale or mean values are specified, a final input blob is computed as: \(input(n,c,h,w) = scalefactor \times (blob(n,c,h,w) - mean_c)\)

    • setInput

      public void setInput(Mat blob, String name, double scalefactor)
      Sets the new input value for the network
      Parameters:
      blob - A new blob. Should have CV_32F or CV_8U depth.
      name - A name of input layer.
      scalefactor - An optional normalization scale. SEE: connect(String, String) to know format of the descriptor. If scale or mean values are specified, a final input blob is computed as: \(input(n,c,h,w) = scalefactor \times (blob(n,c,h,w) - mean_c)\)

    • setInput

      public void setInput(Mat blob, String name)
      Sets the new input value for the network
      Parameters:
      blob - A new blob. Should have CV_32F or CV_8U depth.
      name - A name of input layer. SEE: connect(String, String) to know format of the descriptor. If scale or mean values are specified, a final input blob is computed as: \(input(n,c,h,w) = scalefactor \times (blob(n,c,h,w) - mean_c)\)

    • setInput

      public void setInput(Mat blob)
      Sets the new input value for the network
      Parameters:
      blob - A new blob. Should have CV_32F or CV_8U depth. SEE: connect(String, String) to know format of the descriptor. If scale or mean values are specified, a final input blob is computed as: \(input(n,c,h,w) = scalefactor \times (blob(n,c,h,w) - mean_c)\)

    • setParam

      public void setParam(int layer, int numParam, Mat blob)
      Sets the new value for the learned param of the layer.
      Parameters:
      layer - name or id of the layer.
      numParam - index of the layer parameter in the Layer::blobs array.
      blob - the new value. SEE: Layer::blobs Note: If shape of the new blob differs from the previous shape, then the following forward pass may fail.

    • setParam

      public void setParam(String layerName, int numParam, Mat blob)
      Sets the parameter blob of a layer identified by its name or output tensor name.
      Parameters:
      layerName - layer name (classic engine) or raw ONNX output tensor name (ENGINE_NEW).
      numParam - index of the constant weight input to update (0 = kernel, 1 = bias, etc.).
      blob - the new parameter value.

    • getParam

      public Mat getParam(int layer, int numParam)
      Returns parameter blob of the layer.
      Parameters:
      layer - name or id of the layer.
      numParam - index of the layer parameter in the Layer::blobs array. SEE: Layer::blobs
      Returns:
      automatically generated

    • getParam

      public Mat getParam(int layer)
      Returns parameter blob of the layer.
      Parameters:
      layer - name or id of the layer. SEE: Layer::blobs
      Returns:
      automatically generated

    • getParam

      public Mat getParam(String layerName, int numParam)

    • getParam

      public Mat getParam(String layerName)

    • getUnconnectedOutLayers

      public MatOfInt getUnconnectedOutLayers()
      Returns indexes of layers with unconnected outputs. FIXIT: Rework API to registerOutput() approach, deprecate this call
      Returns:
      automatically generated

    • getUnconnectedOutLayersNames

      public List<String> getUnconnectedOutLayersNames()
      Returns names of layers with unconnected outputs. FIXIT: Rework API to registerOutput() approach, deprecate this call
      Returns:
      automatically generated

    • getLayerShapes

      public void getLayerShapes(List<MatOfInt> netInputShapes, MatOfInt netInputTypes, int layerId, List<MatOfInt> inLayerShapes, List<MatOfInt> outLayerShapes)
      The only overload of getLayerShapes that should be kept in 5.x
      Parameters:
      netInputShapes - automatically generated
      netInputTypes - automatically generated
      layerId - automatically generated
      inLayerShapes - automatically generated
      outLayerShapes - automatically generated

    • getFLOPS

      public long getFLOPS(List<MatOfInt> netInputShapes, MatOfInt netInputTypes)
      Computes FLOP for whole loaded model with specified input shapes.
      Parameters:
      netInputShapes - vector of shapes for all net inputs.
      netInputTypes - vector of types for all net inputs.
      Returns:
      computed FLOP.

    • getLayerTypes

      public void getLayerTypes(List<String> layersTypes)
      Returns list of types for layer used in model.
      Parameters:
      layersTypes - output parameter for returning types.

    • getLayersCount

      public int getLayersCount(String layerType)
      Returns count of layers of specified type.
      Parameters:
      layerType - type.
      Returns:
      count of layers

    • getMemoryConsumption

      public void getMemoryConsumption(List<MatOfInt> netInputShapes, MatOfInt netInputTypes, long[] weights, long[] blobs)
      Computes bytes number which are required to store all weights and intermediate blobs for model.
      Parameters:
      netInputShapes - vector of shapes for all net inputs.
      netInputTypes - vector of types for all net inputs.
      weights - output parameter to store resulting bytes for weights.
      blobs - output parameter to store resulting bytes for intermediate blobs.

    • enableFusion

      public void enableFusion(boolean fusion)
      Enables or disables layer fusion in the network.
      Parameters:
      fusion - true to enable the fusion, false to disable. The fusion is enabled by default.

    • enableWinograd

      public void enableWinograd(boolean useWinograd)
      Enables or disables the Winograd compute branch. The Winograd compute branch can speed up 3x3 Convolution at a small loss of accuracy.
      Parameters:
      useWinograd - true to enable the Winograd compute branch. The default is true.

    • getPerfProfile

      public long getPerfProfile(MatOfDouble timings)
      Returns overall time for inference and timings (in ticks) for layers. Indexes in returned vector correspond to layers ids. Some layers can be fused with others, in this case zero ticks count will be return for that skipped layers. Supported by DNN_BACKEND_OPENCV on DNN_TARGET_CPU only.
      Parameters:
      timings - vector for tick timings for all layers.
      Returns:
      overall ticks for model inference.

    • enableKVCache

      public void enableKVCache()
      Enables KV-Cache for all AttentionOnnxI layers

    • disableKVCache

      public void disableKVCache()
      Disables KV-Cache for all AttentionOnnxI layers

    • resetKVCache

      public void resetKVCache()
      Resets KV-Cache for all AttentionOnnxI layers

    • getPerfProfile

      public void getPerfProfile(List<String> names, List<String> timems, List<String> counts)
      Returns profiling data captured during the last forward pass. Entries are sorted by time in descending order. Empty vectors are returned if profiling is disabled (DNN_PROFILE_NONE).
      Parameters:
      names - automatically generated
      timems - automatically generated
      counts - automatically generated

    • printPerfProfile

      public void printPerfProfile()
      Prints the profile captured during the last forward pass in a formatted table using CV_LOG_INFO. In DNN_PROFILE_DETAILED mode, prints per-layer label, time, and percentage. In DNN_PROFILE_SUMMARY mode, prints per-type count, time, and percentage. Does nothing if profiling is disabled (DNN_PROFILE_NONE) or all timings are zero.