Mulitple Output classification on Nano

kwstastzoumas · September 25, 2020, 2:00pm

Hi,
i would like know if there is any ready-to-run example/project with a two-outputs (two output braches ) caffe-trained classification model running on nano with TensorRT ! I have successfully created the TRT engine for my model (with outputs “prob1” and “prob2”) but i cannot ported to the pre-compiled Googlenet example from jetson-inference repository since this example uses a pre-complied “trtNet” that only has supports a single classification output… Do i need to alter the trtNet.cpp and compile the whole project from scrtach? Any points on what to do as a quick proof of concept ?
thanx, any how!

AastaLLL · September 28, 2020, 4:59am

Hi,

It’s recommended to check detection related sample since it usually has multiple output for location and confidence.
For example:

github.com

dusty-nv/jetson-inference/blob/master/c/detectNet.cpp#L755


      
          	

          	const float cell_width  = /*width*/ GetInputWidth() / ow;

          	const float cell_height = /*height*/ GetInputHeight() / oh;

          	

          	const float scale_x = float(width) / float(GetInputWidth());

          	const float scale_y = float(height) / float(GetInputHeight());

          

          #ifdef DEBUG_CLUSTERING	

          	LogDebug(LOG_TRT "input width %u height %u\n", GetInputWidth(), GetInputHeight());

          	LogDebug(LOG_TRT "cells x %i  y %i\n", ow, oh);

          	LogDebug(LOG_TRT "cell width %f  height %f\n", cell_width, cell_height);

          	LogDebug(LOG_TRT "scale x %f  y %f\n", scale_x, scale_y);

          #endif

          

          	// extract and cluster the raw bounding boxes that meet the coverage threshold

          	int numDetections = 0;

          

          	for( uint32_t z=0; z < cls; z++ )	// z = current object class

          	{

          		for( uint32_t y=0; y < oh; y++ )

          		{

Thanks.

kwstastzoumas · September 29, 2020, 9:26am

thanx! i will try that !

Topic		Replies	Views
Multiple inputs/outputs in TensorRT TensorRT	2	4223	October 12, 2021
Tensorrt output NAN for FP16 and for FP16+INT8 TensorRT tensorrt , jetson-inference , cudnn , jetson	0	37	November 15, 2024
Realize prediction with .trt model (Jetson Nano 2gb) TensorRT	2	569	April 6, 2022
In jetson nano,using my own mobilenetv2 model Jetson Nano jetson-inference	4	464	October 15, 2021
How to use .trt file for inference on jetson nano Jetson Nano tensorrt	4	1533	October 18, 2021
Jetson nano tensorflow-keras cnn model TensorRT	1	1015	January 17, 2022
Running two models in parallel on jetson nano Jetson Nano	5	1768	October 14, 2021
Multiple inference on Jetson Nano Jetson Nano	6	2040	October 18, 2021
Classification TF_2 QAT - Calibration files? TAO Toolkit	9	377	June 4, 2023
New repository for collecting object detection benchmark speed results + examples Jetson Nano tensorrt , ssd , yolo	7	1101	October 15, 2021

Mulitple Output classification on Nano

Related topics