I’m trying to export and convert
.tlt model into
.etlt and tensorrt engine.
I wish to deploy the model onto jetson device,
so I followed the instruction in jupyter notebook (yolo_v4) as follows.
Verify engine generation using the
tao-converterutility included with the docker.
tao-converterproduces optimized tensorrt engines for the platform that it resides on. Therefore, to get maximum performance, please instantiate this docker and execute the
tao-convertercommand, with the exported
.etltfile and calibration cache (for int8 mode) on your target device. The tao-converter utility included in this docker only works for x86 devices, with discrete NVIDIA GPU’s.
For the jetson devices, please download the tao-converter for jetson from the dev zone link here.
The -p argument in following command is the optimization profile. This should be in format
<input_node>,<min_shape>,<opimal_shape>,<max_shape>. In YOLOv3, the three shapes should only have differences at the batch dimension
It leads me to download
clara_agx-20210820T231311Z-001 file, and its help message says, it requires
-o flag for exporting.
But I have no idea what I should type in.
Where/How can I get this information?