First of all I am fairly new to the field so please bare with my questions if they seem trivial. The implementation of YoloV4 I am using is based on keras from this repo . Because of licensing I cannot use Darknet based implementation.
I would like to convert my trained model to tensorRT and infer from it. Is there a way I can utilise yolo to onnx and onnx to tensort that you have implemented ?
This is not guaranteed to work. But you might try to convert your Keras model to ONNX first, then use "onnx_to_tensorrt.py" to optimize it into a TensorRT engine.
First of all I am fairly new to the field so please bare with my questions if they seem trivial. The implementation of YoloV4 I am using is based on keras from this repo . Because of licensing I cannot use Darknet based implementation.
I would like to convert my trained model to tensorRT and infer from it. Is there a way I can utilise yolo to onnx and onnx to tensort that you have implemented ?