create_dl_layer_matmulT_create_dl_layer_matmulCreateDlLayerMatmulCreateDlLayerMatmulcreate_dl_layer_matmul (Operator)

Name

create_dl_layer_matmulT_create_dl_layer_matmulCreateDlLayerMatmulCreateDlLayerMatmulcreate_dl_layer_matmul — Create a MatMul layer.

Signature

Description

The operator create_dl_layer_matmulcreate_dl_layer_matmulCreateDlLayerMatmulCreateDlLayerMatmulcreate_dl_layer_matmul creates a MatMul layer whose handle is returned in DLLayerMatMulDLLayerMatMulDLLayerMatMulDLLayerMatMuldllayer_mat_mul.

A MatMul layer multiplies the 2D matrices, given in the latter two dimensions (H, W) of input DLLayerADLLayerADLLayerADLLayerAdllayer_a, with the corresponding 2D matrices of input DLLayerBDLLayerBDLLayerBDLLayerBdllayer_b, also given in the latter two dimensions (H, W). The output in DLLayerMatMulDLLayerMatMulDLLayerMatMulDLLayerMatMuldllayer_mat_mul is hence given by .

The MatMul layer supports broadcasting for the first input DLLayerADLLayerADLLayerADLLayerAdllayer_a. That means, if the batch size or the number of channels in DLLayerADLLayerADLLayerADLLayerAdllayer_a equals one then the first batch item or channel of DLLayerADLLayerADLLayerADLLayerAdllayer_a is multiplied with all batch items or channels of DLLayerBDLLayerBDLLayerBDLLayerBdllayer_b, respectively.

To make the multiplication work, the width of DLLayerADLLayerADLLayerADLLayerAdllayer_a must be equal to the height of DLLayerBDLLayerBDLLayerBDLLayerBdllayer_b.

The following generic parameters GenParamNameGenParamNameGenParamNamegenParamNamegen_param_name and the corresponding values GenParamValueGenParamValueGenParamValuegenParamValuegen_param_value are supported:

'is_inference_output'"is_inference_output""is_inference_output""is_inference_output""is_inference_output":

Determines whether apply_dl_modelapply_dl_modelApplyDlModelApplyDlModelapply_dl_model will include the output of this layer in the dictionary DLResultBatchDLResultBatchDLResultBatchDLResultBatchdlresult_batch even without specifying this layer in OutputsOutputsOutputsoutputsoutputs ('true'"true""true""true""true") or not ('false'"false""false""false""false").

Default: 'false'"false""false""false""false"

'num_trainable_params'"num_trainable_params""num_trainable_params""num_trainable_params""num_trainable_params":

Number of trainable parameters (weights and biases) of the layer.

'transpose_a'"transpose_a""transpose_a""transpose_a""transpose_a":

Matrices of input DLLayerADLLayerADLLayerADLLayerAdllayer_a are transposed: .

Default: 'false'"false""false""false""false"

'transpose_b'"transpose_b""transpose_b""transpose_b""transpose_b":

Matrices of input DLLayerBDLLayerBDLLayerBDLLayerBdllayer_b are transposed: .

Default: 'false'"false""false""false""false"

Certain parameters of layers created using this operator create_dl_layer_matmulcreate_dl_layer_matmulCreateDlLayerMatmulCreateDlLayerMatmulcreate_dl_layer_matmul can be set and retrieved using further operators. The following tables give an overview, which parameters can be set using set_dl_model_layer_paramset_dl_model_layer_paramSetDlModelLayerParamSetDlModelLayerParamset_dl_model_layer_param and which ones can be retrieved using get_dl_model_layer_paramget_dl_model_layer_paramGetDlModelLayerParamGetDlModelLayerParamget_dl_model_layer_param or get_dl_layer_paramget_dl_layer_paramGetDlLayerParamGetDlLayerParamget_dl_layer_param. Note, the operators set_dl_model_layer_paramset_dl_model_layer_paramSetDlModelLayerParamSetDlModelLayerParamset_dl_model_layer_param and get_dl_model_layer_paramget_dl_model_layer_paramGetDlModelLayerParamGetDlModelLayerParamget_dl_model_layer_param require a model created by create_dl_modelcreate_dl_modelCreateDlModelCreateDlModelcreate_dl_model.

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

Module

Deep Learning Training