Auto tuning

Description

Use auto tuning for cuDNN library.

Category: Code Generation > Interface

Settings

Default: On

On

Use this option to enable auto tuning feature. Enabling auto tuning allows the cuDNN library to find the fastest convolution algorithms. This increases performance for larger networks such as SegNet and ResNet.

Off

Disable auto tuning for the cuDNN library.

Dependencies

• This parameter requires a GPU Coder™ license.

• To enable this parameter, select C++ for the Language and cuDNN for Target library.

Command-Line Information

Parameter: DLAutoTuning

Value: 'on' | 'off'

Default: 'on'

Related Topics

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