Target library

Description

Specify the target deep learning library used during code generation.

Category: Code Generation > Interface

Settings

Default: None

MKL-DNN

Use this option for generating code that uses the Intel^® Math Kernel Library for Deep Neural Networks (Intel MKL-DNN).

ARM Compute

Use this option for generating code that uses the ARM^® Compute Library.

cuDNN

Use this option for generating code that uses the CUDA^® Deep Neural Network library (cuDNN).

TensorRT

Use this option for generating code that takes advantage of the NVIDIA^® TensorRT – high performance deep learning inference optimizer and run-time library.

Dependencies

• To enable this parameter, select C++ for the Language and grt.tlc or ert.tlc for System target file.

• MKL-DNN or ARM Compute is available when GPU acceleration on the Code Generation pane is disabled.

• cuDNN or TensorRT requires a GPU Coder™ license.

• cuDNN or TensorRT is available when GPU acceleration on the Code Generation pane is enabled.

• When the Target library is MKL-DNN, the Language standard parameter on the Code Generation pane is set to C++11 (ISO).

Command-Line Information

Parameter: DLTargetLib

Type: character vector

Value: 'None' | 'MKL-DNN' | 'ARM Compute' | 'cuDNN' | 'TensorRT'

Default: 'None'

Related Topics

• Model Configuration Parameters: Code Generation Interface
• GPU Code Generation for Deep Learning Networks Using MATLAB Function Block (GPU Coder)