1. NPU¶
RK3576 has a NPU(Neural Process Unit) that Neural network acceleration engine with processing performance up to 6 TOPS. Using this NPU module needs to download RKNN SDK which provides programming interfaces for RK3588S/RK3588 chip platforms with NPU. This SDK can help users deploy RKNN models exported by RKNN-Toolkit2 and accelerate the implementation of AI applications
1.1. RKNN Model¶
RKNN is the model type used by the Rockchip NPU platform. It is a model file ending with the suffix .rknn . RKNN SDK provides a complete model transformation Python tool for users to convert their self-developed algorithm model into RKNN model
The RKNN model can run directly on the RK3576 platform. There are demos under rknpu2/examples. Refer to the README.md to compile Android or Linux Demo (Need cross-compile environment). You can also just download compiled demo.
Run demo on the AIO-3576Q as follows:
:/ # cd /data/rknn_ssd_demo_Android/ (use rknn_ssd_demo_Linux in Linux System)
:/data/rknn_ssd_demo_Android # chmod 777 rknn_ssd_demo
:/data/rknn_ssd_demo_Android # export LD_LIBRARY_PATH=./lib
:/data/rknn_ssd_demo_Android # ./rknn_ssd_demo model/RK3588/ssd_inception_v2.rknn model/road.bmp
Loading model ...
rknn_init ...
model input num: 1, output num: 2
input tensors:
index=0, name=Preprocessor/sub:0, n_dims=4, dims=[1, 300, 300, 3], n_elems=270000, size=270000, fmt=NHWC, type=INT8, qnt_type=AFFINE, zp=0, scale=0.007812
output tensors:
index=0, name=concat:0, n_dims=4, dims=[1, 1917, 1, 4], n_elems=7668, size=7668, fmt=NCHW, type=INT8, qnt_type=AFFINE, zp=53, scale=0.089455
index=1, name=concat_1:0, n_dims=4, dims=[1, 1917, 91, 1], n_elems=174447, size=174447, fmt=NCHW, type=INT8, qnt_type=AFFINE, zp=53, scale=0.143593
rknn_run
loadLabelName
ssd - loadLabelName ./model/coco_labels_list.txt
loadBoxPriors
person @ (12 124 60 209) 0.982374
person @ (110 119 153 195) 0.973138
bicycle @ (172 163 277 233) 0.973138
person @ (207 113 258 218) 0.969119
car @ (148 133 215 170) 0.953267
person @ (83 134 92 158) 0.639771
person @ (49 133 58 156) 0.606060
person @ (96 134 105 162) 0.428693
1.2. Non-RKNN Model¶
For other models like Caffe, TensorFlow, etc, to run on RK3576 platform, conversions are needed. Use RKNN-Toolkit2 to convert other model into RKNN model.
1.3. RKNN-Toolkit2¶
1.3.1. Introduction of Tool¶
RKNN-Toolkit2 is a development kit that provides users with model conversion, inference and performance evaluation on PC and Rockchip NPU platforms. Users can easily complete the following functions through the Python interface provided by the tool:
Model conversion: support to convert
Caffe/TensorFlow/TensorFlow Lite/ONNX/Darknet/PyTorchmodel to RKNN model, support RKNN model import/export, which can be used on Rockchip NPU platform laterQuantization: support to convert float model to quantization model, currently support quantized methods including asymmetric quantization(asymmetric_quantized-8, asymmetric_quantized-16). and support hybrid quantization. Asymmetric_quantized-16 not supported yet
Model inference: Able to simulate Rockchip NPU to run RKNN model on PC and get the inference result. This tool can also distribute the RKNN model to the specified NPU device to run, and get the inference results
Performance evaluation: distribute the RKNN model to the specified NPU device to run, and evaluate the model performance in the actual device
Memory evaluation: Evaluate memory consumption at runtime of the model. When using this function, the RKNN model must be distributed to the NPU device to run, and then call the relevant interface to obtain memory information
Quantitative error analysis: This function will give the Euclidean or cosine distance of each layer of inference results before and after the model is quantized. This can be used to analyze how quantitative error occurs, and provide ideas for improving the accuracy of quantitative models
1.3.2. Environment Dependence¶
The system needs: Ubuntu 18.04 (x64) or later. The Toolkit can only be installed on PC, and Windows, MacOS or Debian not supported yet
Python version: 3.6/3.8
Python rely on：
#Python3.6
cat doc/requirements_cp36-1.3.0.txt
numpy==1.16.6
onnx==1.7.0
onnxoptimizer==0.1.0
onnxruntime==1.6.0
tensorflow==1.14.0
tensorboard==1.14.0
protobuf==3.12.0
torch==1.6.0
torchvision==0.7.0
psutil==5.6.2
ruamel.yaml==0.15.81
scipy==1.2.1
tqdm==4.27.0
requests==2.21.0
opencv-python==4.4.0.46
PuLP==2.4
scikit_image==0.17.2
# if install bfloat16 failed, please install numpy manually first. "pip install numpy==1.16.6"
bfloat16==1.1
flatbuffers==1.12
#Python3.8
cat doc/requirements_cp38-1.3.0.txt
numpy==1.17.3
onnx==1.7.0
onnxoptimizer==0.1.0
onnxruntime==1.6.0
tensorflow==2.2.0
tensorboard==2.2.2
protobuf==3.12.0
torch==1.6.0
torchvision==0.7.0
psutil==5.6.2
ruamel.yaml==0.15.81
scipy==1.4.1
tqdm==4.27.0
requests==2.21.0
opencv-python==4.4.0.46
PuLP==2.4
scikit_image==0.17.2
# if install bfloat16 failed, please install numpy manually first. "pip install numpy==1.17.3"
bfloat16==1.1
1.3.3. RKNN-Toolkit2 installation¶
It is recommended to use virtualenv to manage the python environment because there may be multiple versions of the python environment in the system at the same time. Let’s start with Python 3.6 as an example:
# 1）Install virtualenv、Python3.6 and pip3
sudo apt install virtualenv \
sudo apt-get install python3 python3-dev python3-pip
# 2）Install dependent libraries
sudo apt-get install libxslt1-dev zlib1g zlib1g-dev libglib2.0-0 libsm6 \
libgl1-mesa-glx libprotobuf-dev gcc
# 3）Use virtualenv and install Python dependency, such as requirements_cp36-1.3.0.txt
virtualenv -p /usr/bin/python3 venv
source venv/bin/activate
pip3 install -r doc/requirements_cp36-*.txt
# 4）Install RKNN-Toolkit2，such as rknn_toolkit2-1.3.0_11912b58-cp36-cp36m-linux_x86_64.whl
sudo pip3 install packages/rknn_toolkit2*cp36*.whl
# 5）Check if RKNN-Toolkit2 is installed successfully or not，press key ctrl+d to exit
(venv) firefly@T-chip:~/rknn-toolkit2$ python3
>>> from rknn.api import RKNN
>>>
The installation is successful if the import of RKNN module doesn’t fail. One of the failures is as follows:
>>> from rknn.api import RKNN
Traceback (most recent call last):
File "<stdin>", line 1, in <module>
ImportError: No module named 'rknn'
1.3.4. Model Conversion Demo¶
Toolkit Demos are under rknn-toolkit2/examples. Here we run a model conversion demo for example, this demo shows the process of converting tflite to RKNN, exporting model, inferencing, deploying on NPU and fetching results. For detailed implementation of the model conversion, please refer to the source code in Demo and the documents at the end of this page.
1.3.4.1. Simulate the running example on PC¶
RKNN-Toolkit2 has a built-in simulator, simply run a demo to deploy on NPU simulator.
(venv) firefly@T-chip:~/rknn-toolkit2$ cd examples/tflite/mobilenet_v1
(venv) firefly@T-chip:~/rknn-toolkit2/examples/tflite/mobilenet_v1$ ls
dataset.txt dog_224x224.jpg mobilenet_v1_1.0_224.tflite test.py
(venv) firefly@T-chip:~/rknn-toolkit2/examples/tflite/mobilenet_v1$ python3 test.py
W __init__: rknn-toolkit2 version: 1.3.0-11912b58
--> Config model
W config: 'target_platform' is None, use rk3566 as default, Please set according to the actual platform!
done
--> Loading model
INFO: Initialized TensorFlow Lite runtime.
done
--> Building model
Analysing : 100%|█████████████████████████████████████████████████| 58/58 [00:00<00:00, 1869.33it/s]
Quantizating : 100%|████████████████████████████████████████████████| 58/58 [00:00<00:00, 68.07it/s]
W build: The default input dtype of 'input' is changed from 'float32' to 'int8' in rknn model for performance!
Please take care of this change when deploy rknn model with Runtime API!
done
--> Export rknn model
done
--> Init runtime environment
Analysing : 100%|█████████████████████████████████████████████████| 60/60 [00:00<00:00, 1434.93it/s]
Preparing : 100%|██████████████████████████████████████████████████| 60/60 [00:00<00:00, 373.17it/s]
W init_runtime: target is None, use simulator!
done
--> Running model
mobilenet_v1
-----TOP 5-----
[156]: 0.9345703125
[155]: 0.0570068359375
[205]: 0.00429534912109375
[284]: 0.003116607666015625
[285]: 0.00017178058624267578
done
1.3.4.2. Run on AIO-3576Q NPU connected to the PC¶
RKNN Toolkit2 runs on the PC and connects to the AIO-3576Q through the PC’s USB. RKNN Toolkit2 transfers the RKNN model to the NPU device of AIO-3576Q to run, and then obtains the inference results, performance information, etc. from the AIO-3576Q
First prepare AIO-3576Q environment: update librknnrt.so and run rknn_server
Android
adb root && adb remount
adb push rknpu2/runtime/RK3588/Android/rknn_server/arm64-v8a/vendor/bin/rknn_server /vendor/bin
adb push rknpu2/runtime/RK3588/Android/librknn_api/arm64-v8a/librknnrt.so /vendor/lib64
adb push rknpu2/runtime/RK3588/Android/librknn_api/arm64-v8a/librknnrt.so /vendor/lib
adb shell reboot
# Android System will automatically run rknn_server, chekc it by using command "ps -ef | grep rknn_server"
Linux
adb push rknpu2/runtime/RK3588/Linux/rknn_server/aarch64/usr/bin/rknn_server /usr/bin/
adb push rknpu2/runtime/RK3588/Linux/librknn_api/aarch64/librknnrt.so /usr/lib/
adb push rknpu2/runtime/RK3588/Linux/librknn_api/aarch64/librknn_api.so /usr/lib/
# We can use "systemctl status rknn_server" to check whether the rknn_server is running.
# Without running rknn_server, run it on the serial terminal
chmod +x /usr/bin/rknn_server
/usr/bin/rknn_server
Then modify the demo file
examples/tflite/mobilenet_v1/test.pyon PC, add the target platform in it.
diff --git a/examples/tflite/mobilenet_v1/test.py b/examples/tflite/mobilenet_v1/test.py
index 0507edb..fd2e070 100755
--- a/examples/tflite/mobilenet_v1/test.py
+++ b/examples/tflite/mobilenet_v1/test.py
@@ -24,11 +24,11 @@ def show_outputs(outputs):
if __name__ == '__main__':
# Create RKNN object
- rknn = RKNN(verbose=True)
+ rknn = RKNN()
# Pre-process config
print('--> Config model')
- rknn.config(mean_values=[128, 128, 128], std_values=[128, 128, 128])
+ rknn.config(mean_values=[128, 128, 128], std_values=[128, 128, 128], target_platform='rk3576')
print('done')
# Load model
@@ -62,7 +62,7 @@ if __name__ == '__main__':
# Init runtime environment
print('--> Init runtime environment')
- ret = rknn.init_runtime()
+ ret = rknn.init_runtime(target='rk3576')
if ret != 0:
print('Init runtime environment failed!')
exit(ret)
Run test.py on host PC
(venv) firefly@T-chip:~/rknn-toolkit2/examples/tflite/mobilenet_v1$ python3 test.py
W __init__: rknn-toolkit2 version: 1.3.0-11912b58
--> Config model
done
--> Loading model
INFO: Initialized TensorFlow Lite runtime.
done
--> Building model
Analysing : 100%|█████████████████████████████████████████████████| 58/58 [00:00<00:00, 1903.99it/s]
Quantizating : 100%|███████████████████████████████████████████████| 58/58 [00:00<00:00, 114.55it/s]
W build: The default input dtype of 'input' is changed from 'float32' to 'int8' in rknn model for performance!
Please take care of this change when deploy rknn model with Runtime API!
done
--> Export rknn model
done
--> Init runtime environment
I NPUTransfer: Starting NPU Transfer Client, Transfer version 2.1.0 (b5861e7@2020-11-23T11:50:36)
D RKNNAPI: ==============================================
D RKNNAPI: RKNN VERSION:
D RKNNAPI: API: 1.3.0 (121b661 build: 2022-04-29 11:07:20)
D RKNNAPI: DRV: rknn_server: 1.3.0 (121b661 build: 2022-04-29 11:11:57)
D RKNNAPI: DRV: rknnrt: 1.3.0 (c193be371@2022-05-04T20:16:22)
D RKNNAPI: ==============================================
done
--> Running model
mobilenet_v1
-----TOP 5-----
[156]: 0.93505859375
[155]: 0.057037353515625
[205]: 0.0038814544677734375
[284]: 0.0031185150146484375
[285]: 0.00017189979553222656
done
1.3.5. Other Toolkit Demo¶
Other Toolkit demos can be found under rknn-toolkit2/examples/, such as quantization, accuracy analysis demos. For detailed implementation, please refer to the source code in Demo and the detailed development documents.
1.4. Detailed Development Documents¶
Please refer to <<Rockchip_RKNPU_User_Guide_RKNN_API_*.pdf>> and <<Rockchip_User_Guide_RKNN_Toolkit2_*.pdf>> in RKNN SDK for development.