git clone https://github.com/Daipuwei/Mini-VGG-CIFAR10

The SDK only supports python3.6 or python3.8, here is an example of creating a virtual environment for python3.8.

Install python packages.

$ sudo apt update
$ sudo apt install python3-dev python3-numpy

Follow this docs to install conda.

Then create a virtual environment.

$ conda create -n npu-env python=3.8
$ conda activate npu-env     #activate
$ conda deactivate           #deactivate

Download Tool from rockchip-linux/rknn-toolkit2.

$ git clone https://github.com/rockchip-linux/rknn-toolkit2.git
$ git checkout 9ad79343fae625f4910242e370035fcbc40cc31a

Install dependences and RKNN toolkit2 packages.

$ cd rknn-toolkit2
$ sudo apt-get install python3 python3-dev python3-pip
$ sudo apt-get install libxslt1-dev zlib1g-dev libglib2.0 libsm6 libgl1-mesa-glx libprotobuf-dev gcc cmake
$ pip3 install -r doc/requirements_cp38-*.txt
$ pip3 install packages/rknn_toolkit2-*-cp38-cp38-linux_x86_64.whl

Keras model can convert rknn model directly. But this demo is to convert tensorflow model(.pb) first, and convert tnesorflow model to rknn model. We use this tool to convert.

git clone https://github.com/amir-abdi/keras_to_tensorflow.git

Enter rknn-toolkit2/examples/tensorflow/ssd_mobilenet_v1 and modify test.py as follows.

test.py

# Create RKNN object
rknn = RKNN(verbose=True)
 
# Pre-process config
print('--> Config model')
rknn.config(mean_values=[0.0, 0.0, 0.0], std_values=[255.0, 255.0, 255.0], target_platform='rk3588')
print('done')
 
# Load model
print('--> Loading model')
ret = rknn.load_tensorflow(tf_pb='./vgg16.pb',
                            inputs=['image_input'],
                            outputs=['dense_2/Softmax'],
                            input_size_list=[[1, INPUT_SIZE, INPUT_SIZE, 3]])
if ret != 0:
    print('Load model failed!')
    exit(ret)
print('done')
 
# Build Model
print('--> Building model')
ret = rknn.build(do_quantization=True, dataset='./dataset.txt')
if ret != 0:
    print('Build model failed!')
    exit(ret)
print('done')
 
# Export rknn model
print('--> Export rknn model')
ret = rknn.export_rknn('./vgg16.rknn')
if ret != 0:
    print('Export rknn model failed!')
    exit(ret)
print('done')

Run test.py to generate rknn model.

$ python3 test.py

Clone the source code from our khadas/edge2-npu.

$ git clone https://github.com/khadas/edge2-npu

$ sudo apt update
$ sudo apt install cmake libopencv-dev

Put vgg16.rknn in edge2-npu/C++/vgg16/data/model.

# Compile
$ bash build.sh
 
# Run
$ cd install/vgg16
$ ./vgg16 data/model/vgg16.rknn data/img/airplane.jpeg