How to use the tlt exported model to generate an engine in a deepstream environment?

Use the Tansfer Learning Toolkit to train and export the model etlt. In the deepstream3.0 environment, the engine file has been generated. How to use this engine in the deepstream? Have related documents or demos?


You can set the engine file path to the config model-engine-file, and model-file/proto-file are not needed any more.