@spec addLayer(t(), binary(), binary(), integer(), Evision.DNN.LayerParams.t()) ::
  integer() | {:error, String.t()}

Adds new layer to the net.

Positional Arguments

• self: Evision.DNN.Net.t()

• name: String.

  unique name of the adding layer.

• type: String.

  typename of the adding layer (type must be registered in LayerRegister).

• dtype: integer().

  datatype of output blobs.

• params: LayerParams.

  parameters which will be used to initialize the creating layer.

Return

• retval: integer()

@returns unique identifier of created layer, or -1 if a failure will happen.

Python prototype (for reference only):

addLayer(name, type, dtype, params) -> retval

@spec addLayerToPrev(t(), binary(), binary(), integer(), Evision.DNN.LayerParams.t()) ::
  integer() | {:error, String.t()}

Adds new layer and connects its first input to the first output of previously added layer.

Positional Arguments

• self: Evision.DNN.Net.t()
• name: String
• type: String
• dtype: integer()
• params: LayerParams

Return

• retval: integer()

@see addLayer()

Python prototype (for reference only):

addLayerToPrev(name, type, dtype, params) -> retval

@spec connect(t(), binary(), binary()) :: t() | {:error, String.t()}

Connects output of the first layer to input of the second layer.

Positional Arguments

• self: Evision.DNN.Net.t()

• outPin: String.

  descriptor of the first layer output.

• inpPin: String.

  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()

Python prototype (for reference only):

connect(outPin, inpPin) -> None

@spec disableKVCache(Keyword.t()) :: any() | {:error, String.t()}

@spec disableKVCache(t()) :: t() | {:error, String.t()}

Disables KV-Cache for all AttentionOnnxI layers

Positional Arguments

• self: Evision.DNN.Net.t()

Python prototype (for reference only):

disableKVCache() -> None

@spec dump(Keyword.t()) :: any() | {:error, String.t()}

@spec dump(t()) :: binary() | {:error, String.t()}

Dump net to String

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• retval: String

@returns String with structure, hyperparameters, backend, target and fusion Call method after setInput(). To see correct backend, target and fusion run after forward().

Python prototype (for reference only):

dump() -> retval

@spec dumpToFile(t(), binary()) :: t() | {:error, String.t()}

Dump net structure, hyperparameters, backend, target and fusion to dot file

Positional Arguments

• self: Evision.DNN.Net.t()

• path: String.

  path to output file with .dot extension

@see dump()

Python prototype (for reference only):

dumpToFile(path) -> None

@spec dumpToPbtxt(t(), binary()) :: t() | {:error, String.t()}

Dump net structure, hyperparameters, backend, target and fusion to pbtxt file

Positional Arguments

• self: Evision.DNN.Net.t()

• path: String.

  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().

Python prototype (for reference only):

dumpToPbtxt(path) -> None

@spec empty(Keyword.t()) :: any() | {:error, String.t()}

@spec empty(t()) :: boolean() | {:error, String.t()}

empty

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• retval: bool

Returns true if there are no layers in the network.

Python prototype (for reference only):

empty() -> retval

@spec enableFusion(t(), boolean()) :: t() | {:error, String.t()}

Enables or disables layer fusion in the network.

Positional Arguments

• self: Evision.DNN.Net.t()

• fusion: bool.

  true to enable the fusion, false to disable. The fusion is enabled by default.

Python prototype (for reference only):

enableFusion(fusion) -> None

@spec enableKVCache(Keyword.t()) :: any() | {:error, String.t()}

@spec enableKVCache(t()) :: t() | {:error, String.t()}

Enables KV-Cache for all AttentionOnnxI layers

Positional Arguments

• self: Evision.DNN.Net.t()

Python prototype (for reference only):

enableKVCache() -> None

@spec enableWinograd(t(), boolean()) :: t() | {:error, String.t()}

Enables or disables the Winograd compute branch. The Winograd compute branch can speed up 3x3 Convolution at a small loss of accuracy.

Positional Arguments

• self: Evision.DNN.Net.t()

• useWinograd: bool.

  true to enable the Winograd compute branch. The default is true.

Python prototype (for reference only):

enableWinograd(useWinograd) -> None

@spec finalizeNet(Keyword.t()) :: any() | {:error, String.t()}

@spec finalizeNet(t()) :: t() | {:error, String.t()}

Finalizes the network configuration and prepares it for inference.

Positional Arguments

• self: Evision.DNN.Net.t()

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.

Python prototype (for reference only):

finalizeNet() -> None

@spec forward(Evision.Net.t(), [{atom(), term()}, ...] | nil) ::
  [Evision.Mat.t()] | Evision.Mat.t() | {:error, String.t()}

Runs forward pass to compute outputs of layers listed in @p outBlobNames.

Positional Arguments

• self: Evision.DNN.Net.t()

• outBlobNames: [String].

  names for layers which outputs are needed to get

Return

• outputBlobs: [Evision.Mat].

  contains blobs for first outputs of specified layers.

Python prototype (for reference only):

forward(outBlobNames[, outputBlobs]) -> outputBlobs

@spec forwardAndRetrieve(t(), [binary()]) ::
  [[Evision.Mat.t()]] | {:error, String.t()}

Runs forward pass to compute outputs of layers listed in @p outBlobNames.

Positional Arguments

• self: Evision.DNN.Net.t()

• outBlobNames: [String].

  names for layers which outputs are needed to get

Return

• outputBlobs: [[Evision.Mat]].

  contains all output blobs for each layer specified in @p outBlobNames.

Python prototype (for reference only):

forwardAndRetrieve(outBlobNames) -> outputBlobs

@spec forwardAsync(Keyword.t()) :: any() | {:error, String.t()}

@spec forwardAsync(t()) :: Evision.AsyncArray.t() | {:error, String.t()}

Runs forward pass to compute output of layer with name @p outputName.

Positional Arguments

• self: Evision.DNN.Net.t()

Keyword Arguments

• outputName: String.

  name for layer which output is needed to get

Return

• retval: Evision.AsyncArray.t()

@details By default runs forward pass for the whole network. This is an asynchronous version of forward(const String&). dnn::DNN_BACKEND_INFERENCE_ENGINE backend is required.

Python prototype (for reference only):

forwardAsync([, outputName]) -> retval

@spec forwardAsync(t(), [{:outputName, term()}] | nil) ::
  Evision.AsyncArray.t() | {:error, String.t()}

Runs forward pass to compute output of layer with name @p outputName.

Positional Arguments

• self: Evision.DNN.Net.t()

Keyword Arguments

• outputName: String.

  name for layer which output is needed to get

Return

• retval: Evision.AsyncArray.t()

@details By default runs forward pass for the whole network. This is an asynchronous version of forward(const String&). dnn::DNN_BACKEND_INFERENCE_ENGINE backend is required.

Python prototype (for reference only):

forwardAsync([, outputName]) -> retval

@spec getFLOPS(t(), [[integer()]], [integer()]) :: integer() | {:error, String.t()}

Computes FLOP for whole loaded model with specified input shapes.

Positional Arguments

• self: Evision.DNN.Net.t()

• netInputShapes: [Evision.MatShape].

  vector of shapes for all net inputs.

• netInputTypes: [integer()].

  vector of types for all net inputs.

Return

• retval: int64

@returns computed FLOP.

Python prototype (for reference only):

getFLOPS(netInputShapes, netInputTypes) -> retval

@spec getLayer(t(), term()) :: Evision.DNN.Layer.t() | {:error, String.t()}

@spec getLayer(t(), binary()) :: Evision.DNN.Layer.t() | {:error, String.t()}

@spec getLayer(t(), integer()) :: Evision.DNN.Layer.t() | {:error, String.t()}

Variant 1:

getLayer

Positional Arguments

• self: Evision.DNN.Net.t()
• layerId: LayerId

Return

• retval: Evision.DNN.Layer.t()

Has overloading in C++

@deprecated to be removed

Python prototype (for reference only):

getLayer(layerId) -> retval

Variant 2:

getLayer

Positional Arguments

• self: Evision.DNN.Net.t()
• layerName: String

Return

• retval: Evision.DNN.Layer.t()

Has overloading in C++

@deprecated Use int getLayerId(const String &layer)

Python prototype (for reference only):

getLayer(layerName) -> retval

Variant 3:

Returns pointer to layer with specified id or name which the network use.

Positional Arguments

• self: Evision.DNN.Net.t()
• layerId: integer()

Return

• retval: Evision.DNN.Layer.t()

Python prototype (for reference only):

getLayer(layerId) -> retval

@spec getLayerId(t(), binary()) :: integer() | {:error, String.t()}

Converts string name of the layer to the integer identifier.

Positional Arguments

• self: Evision.DNN.Net.t()
• layer: String

Return

• retval: integer()

@returns id of the layer, or -1 if the layer wasn't found.

Python prototype (for reference only):

getLayerId(layer) -> retval

@spec getLayerNames(Keyword.t()) :: any() | {:error, String.t()}

@spec getLayerNames(t()) :: [binary()] | {:error, String.t()}

getLayerNames

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• retval: [String]

Python prototype (for reference only):

getLayerNames() -> retval

@spec getLayerShapes(Evision.Net.t(), [{{atom(), term()}}, ...] | nil) ::
  {[[integer()]], [[integer()]]} | {:error, String.t()}

getLayerShapes

Positional Arguments

• self: Evision.DNN.Net.t()
• netInputShapes: [Evision.MatShape]
• netInputTypes: [integer()]
• layerId: integer()

Return

• inLayerShapes: [Evision.MatShape]
• outLayerShapes: [Evision.MatShape]

Has overloading in C++

The only overload of getLayerShapes that should be kept in 5.x

Python prototype (for reference only):

getLayerShapes(netInputShapes, netInputTypes, layerId) -> inLayerShapes, outLayerShapes

@spec getLayersCount(t(), binary()) :: integer() | {:error, String.t()}

Returns count of layers of specified type.

Positional Arguments

• self: Evision.DNN.Net.t()

• layerType: String.

  type.

Return

• retval: integer()

@returns count of layers

Python prototype (for reference only):

getLayersCount(layerType) -> retval

@spec getLayerTypes(Keyword.t()) :: any() | {:error, String.t()}

@spec getLayerTypes(t()) :: [binary()] | {:error, String.t()}

Returns list of types for layer used in model.

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• layersTypes: [String].

  output parameter for returning types.

Python prototype (for reference only):

getLayerTypes() -> layersTypes

@spec getMemoryConsumption(t(), [[integer()]], [integer()]) ::
  {integer(), integer()} | {:error, String.t()}

Computes bytes number which are required to store all weights and intermediate blobs for model.

Positional Arguments

• self: Evision.DNN.Net.t()

• netInputShapes: [Evision.MatShape].

  vector of shapes for all net inputs.

• netInputTypes: [integer()].

  vector of types for all net inputs.

Return

• weights: size_t.

  output parameter to store resulting bytes for weights.

• blobs: size_t.

  output parameter to store resulting bytes for intermediate blobs.

Python prototype (for reference only):

getMemoryConsumption(netInputShapes, netInputTypes) -> weights, blobs

@spec getModelFormat(Keyword.t()) :: any() | {:error, String.t()}

@spec getModelFormat(t()) :: Evision.DNN.ModelFormat.t() | {:error, String.t()}

Retrieve the current model format, see DNNMODEL*

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• retval: ModelFormat

Python prototype (for reference only):

getModelFormat() -> retval

@spec getParam(t(), binary()) :: Evision.Mat.t() | {:error, String.t()}

@spec getParam(t(), integer()) :: Evision.Mat.t() | {:error, String.t()}

Variant 1:

getParam

Positional Arguments

• self: Evision.DNN.Net.t()
• layerName: String

Keyword Arguments

• numParam: integer().

Return

• retval: Evision.Mat.t()

Python prototype (for reference only):

getParam(layerName[, numParam]) -> retval

Variant 2:

Returns parameter blob of the layer.

Positional Arguments

• self: Evision.DNN.Net.t()

• layer: integer().

  name or id of the layer.

Keyword Arguments

• numParam: integer().

  index of the layer parameter in the Layer::blobs array.

Return

• retval: Evision.Mat.t()

@see Layer::blobs

Python prototype (for reference only):

getParam(layer[, numParam]) -> retval

@spec getParam(t(), binary(), [{:numParam, term()}] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

@spec getParam(t(), integer(), [{:numParam, term()}] | nil) ::
  Evision.Mat.t() | {:error, String.t()}

Variant 1:

getParam

Positional Arguments

• self: Evision.DNN.Net.t()
• layerName: String

Keyword Arguments

• numParam: integer().

Return

• retval: Evision.Mat.t()

Python prototype (for reference only):

getParam(layerName[, numParam]) -> retval

Variant 2:

Returns parameter blob of the layer.

Positional Arguments

• self: Evision.DNN.Net.t()

• layer: integer().

  name or id of the layer.

Keyword Arguments

• numParam: integer().

  index of the layer parameter in the Layer::blobs array.

Return

• retval: Evision.Mat.t()

@see Layer::blobs

Python prototype (for reference only):

getParam(layer[, numParam]) -> retval

@spec getPerfProfile(Keyword.t()) :: any() | {:error, String.t()}

@spec getPerfProfile(t()) ::
  {[binary()], [binary()], [binary()]} | {:error, String.t()}

Returns profiling data captured during the last forward pass.

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• names: [string]
• timems: [string]
• counts: [string]

Entries are sorted by time in descending order. Empty vectors are returned if profiling is disabled (DNN_PROFILE_NONE).

Python prototype (for reference only):

getPerfProfile() -> names, timems, counts

@spec getProfilingMode(Keyword.t()) :: any() | {:error, String.t()}

@spec getProfilingMode(t()) :: Evision.DNN.ProfilingMode.t() | {:error, String.t()}

Retrieve the current profiling mode

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• retval: ProfilingMode

Python prototype (for reference only):

getProfilingMode() -> retval

@spec getTracingMode(Keyword.t()) :: any() | {:error, String.t()}

@spec getTracingMode(t()) :: Evision.DNN.TracingMode.t() | {:error, String.t()}

Retrieve the current tracing mode

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• retval: TracingMode

Python prototype (for reference only):

getTracingMode() -> retval

@spec getUnconnectedOutLayers(Keyword.t()) :: any() | {:error, String.t()}

@spec getUnconnectedOutLayers(t()) :: [integer()] | {:error, String.t()}

Returns indexes of layers with unconnected outputs.

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• retval: [integer()]

FIXIT: Rework API to registerOutput() approach, deprecate this call

Python prototype (for reference only):

getUnconnectedOutLayers() -> retval

@spec getUnconnectedOutLayersNames(Keyword.t()) :: any() | {:error, String.t()}

@spec getUnconnectedOutLayersNames(t()) :: [binary()] | {:error, String.t()}

Returns names of layers with unconnected outputs.

Positional Arguments

• self: Evision.DNN.Net.t()

Return

• retval: [String]

FIXIT: Rework API to registerOutput() approach, deprecate this call

Python prototype (for reference only):

getUnconnectedOutLayersNames() -> retval

@spec net() :: t() | {:error, String.t()}

Net

Return

• self: Evision.DNN.Net.t()

Python prototype (for reference only):

Net() -> <dnn_Net object>

@spec printPerfProfile(Keyword.t()) :: any() | {:error, String.t()}

@spec printPerfProfile(t()) :: t() | {:error, String.t()}

Prints the profile captured during the last forward pass in a formatted table using CV_LOG_INFO.

Positional Arguments

• self: Evision.DNN.Net.t()

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.

Python prototype (for reference only):

printPerfProfile() -> None

@spec readFromModelOptimizer(binary(), binary()) :: t() | {:error, String.t()}

Create a network from Intel's Model Optimizer in-memory buffers with intermediate representation (IR).

Positional Arguments

• bufferModelConfig: [uchar].

  buffer with model's configuration.

• bufferWeights: [uchar].

  buffer with model's trained weights.

Return

• retval: Evision.DNN.Net.t()

@returns Net object.

Python prototype (for reference only):

readFromModelOptimizer(bufferModelConfig, bufferWeights) -> retval

@spec registerOutput(t(), binary(), integer(), integer()) ::
  integer() | {:error, String.t()}

Registers network output with name

Positional Arguments

• self: Evision.DNN.Net.t()

• outputName: string.

  identifier of the output

• layerId: integer().

  identifier of the second layer

• outputPort: integer().

  number of the second layer input

Return

• retval: integer()

  Function may create additional 'Identity' layer.

@returns index of bound layer (the same as layerId or newly created)

Python prototype (for reference only):

registerOutput(outputName, layerId, outputPort) -> retval

@spec resetKVCache(Keyword.t()) :: any() | {:error, String.t()}

@spec resetKVCache(t()) :: t() | {:error, String.t()}

Resets KV-Cache for all AttentionOnnxI layers

Positional Arguments

• self: Evision.DNN.Net.t()

Python prototype (for reference only):

resetKVCache() -> None

@spec setInput(t(), Evision.Mat.maybe_mat_in()) :: t() | {:error, String.t()}

Sets the new input value for the network

Positional Arguments

• self: Evision.DNN.Net.t()

• blob: Evision.Mat.

  A new blob. Should have CV_32F or CV_8U depth.

Keyword Arguments

• name: String.

  A name of input layer.

• scalefactor: double.

  An optional normalization scale.

• mean: Evision.scalar().

  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: \f[input(n,c,h,w) = scalefactor \times (blob(n,c,h,w) - mean_c)\f]

Python prototype (for reference only):

setInput(blob[, name[, scalefactor[, mean]]]) -> None

@spec setInput(
  t(),
  Evision.Mat.maybe_mat_in(),
  [mean: term(), name: term(), scalefactor: term()] | nil
) ::
  t() | {:error, String.t()}

Sets the new input value for the network

Positional Arguments

• self: Evision.DNN.Net.t()

• blob: Evision.Mat.

  A new blob. Should have CV_32F or CV_8U depth.

Keyword Arguments

• name: String.

  A name of input layer.

• scalefactor: double.

  An optional normalization scale.

• mean: Evision.scalar().

  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: \f[input(n,c,h,w) = scalefactor \times (blob(n,c,h,w) - mean_c)\f]

Python prototype (for reference only):

setInput(blob[, name[, scalefactor[, mean]]]) -> None

@spec setInputShape(t(), binary(), [integer()]) :: t() | {:error, String.t()}

Specify shape of network input.

Positional Arguments

• self: Evision.DNN.Net.t()
• inputName: String
• shape: Evision.MatShape

Python prototype (for reference only):

setInputShape(inputName, shape) -> None

@spec setInputsNames(t(), [binary()]) :: t() | {:error, String.t()}

Sets outputs names of the network input pseudo layer.

Positional Arguments

• self: Evision.DNN.Net.t()
• inputBlobNames: [String]

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.

Python prototype (for reference only):

setInputsNames(inputBlobNames) -> None

@spec setParam(t(), binary(), integer(), Evision.Mat.maybe_mat_in()) ::
  t() | {:error, String.t()}

@spec setParam(t(), integer(), integer(), Evision.Mat.maybe_mat_in()) ::
  t() | {:error, String.t()}

Variant 1:

Sets the parameter blob of a layer identified by its name or output tensor name.

Positional Arguments

• self: Evision.DNN.Net.t()

• layerName: String.

  layer name (classic engine) or raw ONNX output tensor name (ENGINE_NEW).

• numParam: integer().

  index of the constant weight input to update (0 = kernel, 1 = bias, etc.).

• blob: Evision.Mat.

  the new parameter value.

Python prototype (for reference only):

setParam(layerName, numParam, blob) -> None

Variant 2:

Sets the new value for the learned param of the layer.

Positional Arguments

• self: Evision.DNN.Net.t()

• layer: integer().

  name or id of the layer.

• numParam: integer().

  index of the layer parameter in the Layer::blobs array.

• blob: Evision.Mat.

  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.

Python prototype (for reference only):

setParam(layer, numParam, blob) -> None

@spec setPreferableBackend(t(), integer()) :: t() | {:error, String.t()}

Ask network to use specific computation backend where it supported.

Positional Arguments

• self: Evision.DNN.Net.t()

• backendId: integer().

  backend identifier.

@see Backend

Python prototype (for reference only):

setPreferableBackend(backendId) -> None

@spec setPreferableTarget(t(), integer()) :: t() | {:error, String.t()}

Ask network to make computations on specific target device.

Positional Arguments

• self: Evision.DNN.Net.t()

• targetId: integer().

  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 | | + | |

Python prototype (for reference only):

setPreferableTarget(targetId) -> None

@spec setProfilingMode(t(), Evision.DNN.ProfilingMode.t()) ::
  t() | {:error, String.t()}

Set the profiling mode

Positional Arguments

• self: Evision.DNN.Net.t()

• profilingMode: ProfilingMode.

  the profiling mode, see DNNPROFILE*

Python prototype (for reference only):

setProfilingMode(profilingMode) -> None

@spec setTracingMode(t(), Evision.DNN.TracingMode.t()) :: t() | {:error, String.t()}

Set the tracing mode

Positional Arguments

• self: Evision.DNN.Net.t()

• tracingMode: TracingMode.

  the tracing mode, see DNNTRACE*

Python prototype (for reference only):

setTracingMode(tracingMode) -> None