Skip to content

set_dl_model_layer_weightsSetDlModelLayerWeightsSetDlModelLayerWeightsset_dl_model_layer_weightsT_set_dl_model_layer_weights🔗

Short description🔗

set_dl_model_layer_weightsSetDlModelLayerWeightsSetDlModelLayerWeightsset_dl_model_layer_weightsT_set_dl_model_layer_weights — Set the weights (or values) of a Deep Learning model layer.

Signature🔗

set_dl_model_layer_weights( image Weights, dl_model DLModelHandle, string LayerName, string WeightsType )void SetDlModelLayerWeights( const HObject& Weights, const HTuple& DLModelHandle, const HTuple& LayerName, const HTuple& WeightsType )static void HOperatorSet.SetDlModelLayerWeights( HObject weights, HTuple DLModelHandle, HTuple layerName, HTuple weightsType )def set_dl_model_layer_weights( weights: HObject, dlmodel_handle: HHandle, layer_name: str, weights_type: str ) -> None

Herror T_set_dl_model_layer_weights( const Hobject Weights, const Htuple DLModelHandle, const Htuple LayerName, const Htuple WeightsType )

void HDlModel::SetDlModelLayerWeights( const HImage& Weights, const HString& LayerName, const HString& WeightsType ) const

void HDlModel::SetDlModelLayerWeights( const HImage& Weights, const char* LayerName, const char* WeightsType ) const

void HDlModel::SetDlModelLayerWeights( const HImage& Weights, const wchar_t* LayerName, const wchar_t* WeightsType ) const (Windows only)

void HDlModel.SetDlModelLayerWeights( HImage weights, string layerName, string weightsType )

Description🔗

The operator set_dl_model_layer_weightsSetDlModelLayerWeights sets for the model DLModelHandleDLModelHandledlmodel_handle the given Weightsweightsweights in the specified LayerNamelayerNamelayer_name.

The parameter WeightsTypeweightsTypeweights_type determines which type of layer values are set. Which values can be set, please refer to the get_dl_model_layer_weightsGetDlModelLayerWeights documentation.

Attention🔗

The operator set_dl_model_layer_weightsSetDlModelLayerWeights is only applicable to self-created networks. For networks delivered by HALCON, the operator does have no impact.

Execution information🔗

Execution information

  • Multithreading type: reentrant (runs in parallel with non-exclusive operators).

  • Multithreading scope: global (may be called from any thread).

  • Processed without parallelization.

Parameters🔗

Weightsweightsweights (input_object) (multichannel-)image(-array) → object (real)HObject (real)HImage (real)HObject (real)Hobject (real)

Input weights.

DLModelHandleDLModelHandledlmodel_handle (input_control) dl_model → (handle)HTuple (HHandle)HDlModel, HTuple (IntPtr)HHandleHtuple (handle)

Handle of the deep learning model.

LayerNamelayerNamelayer_name (input_control) string → (string)HTuple (HString)HTuple (string)strHtuple (char*)

Name of the layer, whose weights are to be set.

WeightsTypeweightsTypeweights_type (input_control) string → (string)HTuple (HString)HTuple (string)strHtuple (char*)

Selected type of layer values to be set.

Default: 'weights'"weights"
List of values: 'batchnorm_mean', 'batchnorm_mean_avg', 'batchnorm_variance', 'batchnorm_variance_avg', 'bias', 'weights'"batchnorm_mean", "batchnorm_mean_avg", "batchnorm_variance", "batchnorm_variance_avg", "bias", "weights"

Example🔗

(HDevelop)

* Create weights for a convolution layer.
gen_image_const (Weights, 'real', 1, 1)
paint_region (Weights, Weights, Weights, 1, 'fill')
gen_empty_obj (WeightsArray)
for Index := 0 to 10 by 1
    concat_obj (WeightsArray, Weights, WeightsArray)
endfor
*
* Input image with rows consisting of 1s to 10s.
gen_image_const (Image, 'real', 10, 10)
for Index := 0 to 9 by 1
    gen_rectangle1 (Rectangle, Index, 0, Index, 9)
    paint_region (Rectangle, Image, Image, Index + 1, 'fill')
endfor
*
* Create a small model network.
create_dl_layer_input ('image', [10, 2, 1], [], [], ImageNode)
create_dl_layer_convolution (ImageNode, 'conv', 1, 1, 2, 11, 1, 'none', \
                             'none', [], [], ConvNode)
create_dl_layer_zoom_factor (ConvNode, 'zoom', 2, 2, 'bilinear', 'true', [],\
                             [], ZoomNode)
create_dl_model (ZoomNode, DLModelHandle)
set_dl_model_param (DLModelHandle, 'runtime', 'cpu')
*
* Set the weights to the convolution layer.
set_dl_model_layer_weights (WeightsArray, DLModelHandle, 'conv', 'weights')

Combinations with other operators🔗

Combinations

Possible predecessors

create_dl_modelCreateDlModel, get_dl_model_layer_weightsGetDlModelLayerWeights

Possible successors

get_dl_model_layer_weightsGetDlModelLayerWeights

Module🔗

This operator uses dynamic licensing (see the ‘Installation Guide’). Which of the following modules is required depends on the specific usage of the operator:

Deep Learning Professional