Run custom CNN with python imageNet class

deepancient · May 19, 2020, 9:12pm

Hello, I have train a custom cnn model with keras and then I have obtained the .uff file and the .engine TensorRT model. Is it possible to load my model with the line net = jetson.inference.imageNet(opt.network) ? Thanks

AastaLLL · May 20, 2020, 2:44am

Hi,

YES. Please check this page for more information.

github.com

dusty-nv/jetson-inference/blob/master/docs/imagenet-example-2.md

<img src="https://github.com/dusty-nv/jetson-inference/raw/master/docs/images/deep-vision-header.jpg" width="100%">
<p align="right"><sup><a href="imagenet-example-python-2.md">Back</a> | <a href="imagenet-camera-2.md">Next</a> | </sup><a href="../README.md#hello-ai-world"><sup>Contents</sup></a>
<br/>
<sup>Image Recognition</sup></p>  

# Coding Your Own Image Recognition Program (C++)
In the previous step, we ran an application that came with the jetson-inference repo.  

Now, we're going to walk through creating a new program from scratch for image recognition called [`my-recognition`](../examples/my-recognition/my-recognition.cpp).  This program will be able to exist as a standalone project outside the repo, hence if you wish to use the jetson-inference library in your own projects and applications, you can follow this example.    

``` cpp
#include <jetson-inference/imageNet.h>
#include <jetson-utils/loadImage.h>

int main( int argc, char** argv )
{
	// a command line argument containing the image filename is expected,
	// so make sure we have at least 2 args (the first arg is the program)
	if( argc < 2 )
	{

This file has been truncated. show original

You will need to update the input/output layer, label and model path.
Thanks.

deepancient · May 20, 2020, 9:24pm

Thanks, I will try that

deepancient · May 21, 2020, 12:47am

I have successfully compile and run the the example of the link, then I have loading my .uff model with:

imageNet* net = imageNet::Create(NULL, “models/my_cnn_model.uff”, NULL, “models/labels.txt”,
“input_1”, “dense_1/Softmax”, 1, TYPE_FP16, DEVICE_GPU, true);

It compiles without error but when run it I have this error:

$ ./run_tensor_rt_engine
[image] loaded ‘imgs/no_defectuosa_1.jpg’ (1995 x 1420, 3 channels)

imageNet – loading classification network model from:
– prototxt (null)
– model models/my_cnn_model.uff
– class_labels models/labels.txt
– input_blob ‘input_1’
– output_blob ‘dense_1/Softmax’
– batch_size 1

[TRT] TensorRT version 5.1.6
[TRT] loading NVIDIA plugins…
[TRT] Plugin Creator registration succeeded - GridAnchor_TRT
[TRT] Plugin Creator registration succeeded - NMS_TRT
[TRT] Plugin Creator registration succeeded - Reorg_TRT
[TRT] Plugin Creator registration succeeded - Region_TRT
[TRT] Plugin Creator registration succeeded - Clip_TRT
[TRT] Plugin Creator registration succeeded - LReLU_TRT
[TRT] Plugin Creator registration succeeded - PriorBox_TRT
[TRT] Plugin Creator registration succeeded - Normalize_TRT
[TRT] Plugin Creator registration succeeded - RPROI_TRT
[TRT] Plugin Creator registration succeeded - BatchedNMS_TRT
[TRT] completed loading NVIDIA plugins.
[TRT] detected model format - UFF (extension ‘.uff’)
[TRT] desired precision specified for GPU: FP16
[TRT] native precisions detected for GPU: FP32, FP16
[TRT] attempting to open engine cache file models/my_cnn_model.uff.1.1.GPU.FP16.engine
[TRT] loading network profile from engine cache… models/my_cnn_model.uff.1.1.GPU.FP16.engine
[TRT] device GPU, models/my_cnn_model.uff loaded
[TRT] Glob Size is 4755780 bytes.
[TRT] Added linear block of size 3686400
[TRT] Added linear block of size 1843200
[TRT] Added linear block of size 230400
[TRT] Added linear block of size 205312
[TRT] Added linear block of size 57856
[TRT] Deserialize required 2944367 microseconds.
[TRT] device GPU, CUDA engine context initialized with 2 bindings
[TRT] binding – index 0
– name ‘input_1’
– type FP32
– in/out INPUT
– # dims 3
– dim #0 3 (SPATIAL)
– dim #1 480 (SPATIAL)
– dim #2 480 (SPATIAL)
[TRT] binding – index 1
– name ‘dense_1/Softmax’
– type FP32
– in/out OUTPUT
– # dims 3
– dim #0 1 (SPATIAL)
– dim #1 1 (SPATIAL)
– dim #2 2 (SPATIAL)
[TRT] binding to input 0 input_1 binding index: 0
[TRT] binding to input 0 input_1 dims (b=1 c=3 h=480 w=480) size=2764800
[TRT] binding to output 0 dense_1/Softmax binding index: 1
[TRT] binding to output 0 dense_1/Softmax dims (b=1 c=1 h=1 w=2) size=8
device GPU, models/my_cnn_model.uff initialized.
[TRT] models/my_cnn_model.uff loaded
imageNet – loaded 2 class info entries
imageNet – didn’t load expected number of class descriptions (2 of 1)
imageNet – failed to load synset class descriptions (2 / 2 of 1)
[TRT] imageNet – failed to initialize.
failed to load image recognition network

My model only has two outputs in a softmax layer so I have created this labels.txt file:
1 pass
2 failed

So I don’t understand why TensorRT return this error:

imageNet – loaded 2 class info entries
imageNet – didn’t load expected number of class descriptions (2 of 1)
imageNet – failed to load synset class descriptions (2 / 2 of 1)

deepancient · May 22, 2020, 8:48pm

Comparing the output of the GoogleNet model with my custom model, I have realice where is the problem. GoogleNet output’s is in NCHW format with (b=1 c=1 h=1000 w=1) shape, where 1000 is the number of units in the softmax layer, but my model has (b=1 c=1 h=1 w=2) shape (since it only has 2 units in the output softmax layer).
So, how can I reshape the output to (b=1 c=1 h=2 w=1) to be compatible with Tensor RT?

AastaLLL · June 3, 2020, 5:33am

Hi,

You don’t need to resize the output but update the class number into 2.
Please check this file for detail:

github.com

dusty-nv/jetson-inference/blob/master/c/imageNet.cpp#L268


      
          		// create from pretrained model

          		net = imageNet::Create(type);

          	}

          

          	if( !net )

          		return NULL;

          

          	// enable layer profiling if desired

          	if( cmdLine.GetFlag("profile") )

          		net->EnableLayerProfiler();

          

          	return net;

          }

          

          	 

          // loadClassInfo

          bool imageNet::loadClassInfo( const char* filename, int expectedClasses )

          {

          	if( !LoadClassLabels(filename, mClassDesc, mClassSynset, expectedClasses) )

          		return false;

Thanks.

Topic		Replies	Views
Not able to run custom DNN on Jetson nano Jetson Nano cudnn	10	796	October 18, 2021
AI examples Jetson Nano - Could not register plugin creator - cache file not found Jetson Nano	4	1368	October 14, 2021
Hello AI World, jetson-inference, Classifying Images with ImageNet: imageNet -- failed to initialize Jetson AGX Orin jetson-inference	5	399	March 12, 2024
How to load models from a custom directory? Jetson Nano tensorrt , python	4	909	August 24, 2022
Resnet 18 Jetson Nano jetson-inference	2	478	May 22, 2023
Speed up inference time on Nano with mxnet Jetson Nano neural-network-framework	9	1778	October 18, 2021
Custom ResNet Jetson Xavier Jetson Xavier NX jetson-inference	12	2981	October 18, 2021
Classifying Images with Imagenet Jetson Nano cudnn	2	80	July 2, 2024
Jetson inference failed to load detectNet model Jetson Nano jetson-inference , cudnn	2	262	February 20, 2024
Detectnet failed to load Resnet50 ONNX model Jetson Nano onnx	2	1029	February 2, 2022

Run custom CNN with python imageNet class

Related topics