Parameters for Deepstream to convert onnx to engine

Accelerated Computing Intelligent Video Analytics DeepStream SDK
,
1

Please provide complete information as applicable to your setup.

• Hardware Platform (Jetson / GPU) Jetson TX2
• DeepStream Version 5.1
• JetPack Version (valid for Jetson only)
• TensorRT Version 7.1.3
• NVIDIA GPU Driver Version (valid for GPU only)
• Issue Type( questions, new requirements, bugs) question

Hello,

I know, that Deepstream can automatically convert my model from .onnx to .engine format, if I didn’t use model-engine-file parameter in config.

While using trtexec to convert my model manually, I can chose build options such as --fp16, inputIOFormats and so on.

What parameters are used by Deepstream to convert model from .onnx to .engine? Does resulting.engine enable fp16 precision?

3

You can use the trtexec --help to check the parameters you need.

4

I don’t want to use trtexec to convert my model. I just want to know, which build options are used by DeepStream, when it converts onnx to engine

5

Please refer to our Guide. These parameters are in the configuration file of nvinfer.

6

This topic was automatically closed 14 days after the last reply. New replies are no longer allowed.