Tensorflow to TensorRT Model Conversion for Retraining the Model

Hi,

We use Jetson-inference’s pre-trained object detection model for our application. But we want to train weights with our site dataset. Would you provide a source for converting a Tensorflow model to TensorRT for Xavier Nx?

Thank you

Hi,

Please convert your model into onnx format with tf2onnx or keras2onnx, and following this sample for TensorRT:
https://docs.nvidia.com/deeplearning/tensorrt/sample-support-guide/index.html#yolov3_onnx

Thanks.

@myagmur did you get a performance boost on the nx when you converted your tf model to a tf_rt model?

you have to convert your model into TensorRT to run it on NX

I did, I am using the process described here: https://docs.nvidia.com/deeplearning/frameworks/tf-trt-user-guide/index.html#worflow-with-savedmodel

I am not seeing any performance improvement.