 Kria™ KV260 Vision AI Starter Kit
|
Customizing the AI Models Used in the Application |
Customizing the AI Models Used in the Application¶
Introduction¶
This document provide an overview of how to customize the NLP SmartVision application to use other AI models other than the default ones.
Prerequisites¶
Other than the three models provided via the keywords “up” and “down” as documented here,
facedetect (densebox_640_360)
object Detect (yolov2_voc_pruned_0_77)
plate Detect (plate_detect)
Customization can be made to use other AMD Vitis™ AI models or retrained model by the users of the same class.
Model Preparation¶
NOTE:
The design currently only supports Vitis AI 2.5.0.
As described in the Hardware Accelerator section, the DPU integrated in the platform uses the B3136 configuration.
The arch.json used to compile the xmodel for B3136 DPU can be obtained by build the accelerator, but if you will not build all from the start, you can save the following code as arch.json:
{
"fingerprint":"0x1000020F6014405"
}
For detailed instructions on obtaining an alternative model from the model zoo or training, pruning, quantizing, and compiling a new model, refer to the Vitis AI User Guide (UG1414).
Next Steps¶
Go back to the KV260 SOM NLP SmartVision Design Start Page