Auto tuning
Enable auto tuning for the cuDNN library to find the fastest convolution algorithms
Model Configuration Pane: Simulation Target
Description
The Auto tuning parameter specifies whether to use auto tuning for cuDNN library. Enabling auto tuning allows the cuDNN library to find the fastest convolution algorithms.
Dependencies
This parameter requires a GPU Coder™ license.
To enable this parameter, select
C++for the Language andcuDNNfor Target library.
Settings
Default: On
OnUse 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
SegNetandResNet.
OffDisable auto tuning for the cuDNN library.
Recommended Settings
| Application | Setting |
|---|---|
| Debugging | No impact |
| Traceability | No impact |
| Efficiency | No impact |
| Safety precaution | No impact |
Programmatic Use
Parameter:
SimDLAutoTuning |
Value:
'on' | 'off' |
Default:
'on' |
Version History
Introduced in R2020b
