Khadas Docs

Amazing Khadas, always amazes you!

User Tools

Site Tools

Home / products / sbc / edge-2l / npu / rknn-sdk-quick-start
products:sbc:edge-2l:npu:rknn-sdk-quick-start

Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
Next revision 		Previous revision
products:sbc:edge-2l:npu:rknn-sdk-quick-start [2026/05/31 22:55]
nick [Install Compilation Tools on the Computer] 		products:sbc:edge-2l:npu:rknn-sdk-quick-start [2026/05/31 23:55] (current)
nick [Run RKNN Python demo]
Line 1: Line 1:
-====== RKNN SDK Quick Start ======+====== RKNN SDK quick start ======
  
 ===== Introduction ===== ===== Introduction =====
 This document provides a detailed introduction for beginners on how to quickly use RKNN-Toolkit2 on a computer to perform model conversion and deploy it to a Edge 2L board using RKNPU2. The examples used in this document are integrated into the [[https://github.com/airockchip/rknn_model_zoo | RKNN Model Zoo]]. This document provides a detailed introduction for beginners on how to quickly use RKNN-Toolkit2 on a computer to perform model conversion and deploy it to a Edge 2L board using RKNPU2. The examples used in this document are integrated into the [[https://github.com/airockchip/rknn_model_zoo | RKNN Model Zoo]].
  
-===== Prepare Development Environment =====+===== Prepare development environment =====
 The version information we are using is as follows:  The version information we are using is as follows: 
-  * python: 3.8+  * Host PC Python: 3.8
   * rknn-toolkit2: 2.3.2   * rknn-toolkit2: 2.3.2
   * rknn model zoo: 2.3.2   * rknn model zoo: 2.3.2
  
-==== Download RKNN Related Repositories ====+==== Download RKNN related repositories - Host PC ====
 It is recommended to create a new directory to store the RKNN repositories. For example, create a folder named ''Projects'' and place the RKNN-Toolkit2 and RKNN Model Zoo repositories in that directory. Refer to the following commands It is recommended to create a new directory to store the RKNN repositories. For example, create a folder named ''Projects'' and place the RKNN-Toolkit2 and RKNN Model Zoo repositories in that directory. Refer to the following commands
  
Line 26: Line 26:
 ``` ```
  
-==== Install the RKNN-Toolkit2 environment on your host PC ====+==== Install the RKNN-Toolkit2 environment on your Host PC ====
  
-=== Install Python ===+=== Install Python - Host PC ===
 If the Python 3.8 environment is not installed on your system, or if there are multiple versions of Python installed, it is recommended to use Miniforge Conda to create a new Python 3.8 environment. If the Python 3.8 environment is not installed on your system, or if there are multiple versions of Python installed, it is recommended to use Miniforge Conda to create a new Python 3.8 environment.
  
-=== Install Miniforge Conda ===+=== Install Miniforge Conda - Host PC ===
  
 Download the Miniforge Conda installer from the following link: Download the Miniforge Conda installer from the following link:
Line 43: Line 43:
 $ bash Miniforge3-Linux-x86_64.sh $ bash Miniforge3-Linux-x86_64.sh
 ``` ```
-=== Create Python Environment Using Miniforge Conda ===+=== Create Python environment using miniforge conda - Host PC ===
 In the terminal window on the computer, execute the following command to switch to the Miniforge conda base In the terminal window on the computer, execute the following command to switch to the Miniforge conda base
 environment: environment:
Line 57: Line 57:
 $ conda activate py38Toolkit2.3.2 $ conda activate py38Toolkit2.3.2
 ``` ```
-=== Install RKNN-Toolkit2 ===+=== Install RKNN-Toolkit2 - Host PC ===
 Install via local wheel packag. Install via local wheel packag.
  
Line 66: Line 66:
 ``` ```
  
-=== Check if the RKNN-Toolkit2 Environment is Installed Successfully ===+=== Check if the RKNN-Toolkit2 environment - Host PC ===
 Switch to Python interactive mode Switch to Python interactive mode
 ```shell ```shell
Line 76: Line 76:
 ``` ```
  
-==== Install Compilation Tools on the Computer ==== +==== Install compilation tools on your host PC ==== 
-=== Android - Install NDK on Android system ===+=== Android - Install NDK for Android system ===
 Note: This section is applicable to development boards with the Android system. If the board is running on the Note: This section is applicable to development boards with the Android system. If the board is running on the
 Linux system, please skip this section.\\ Linux system, please skip this section.\\
Line 90: Line 90:
 ``` ```
  
-==== Linux - Install GCC Cross-Compiler on Linux system ====+==== Linux - Install GCC Cross-Compiler for Linux system ====
 Note: This section is applicable to development boards with the Linux system. If the board is running on the Note: This section is applicable to development boards with the Linux system. If the board is running on the
 Android system, please skip this section.\\ Android system, please skip this section.\\
Line 102: Line 102:
 ``` ```
  
-===== Run Example Programs =====+===== Run example programs =====
  
-==== Prepare Model ====+==== Prepare model ====
 Switch to the ''rknn_model_zoo/examples/yolov5/model'' directory, and execute the ''download_model.sh'' script. This Switch to the ''rknn_model_zoo/examples/yolov5/model'' directory, and execute the ''download_model.sh'' script. This
 script will download an available YOLOv5 ONNX model and store it in the current model directory. Refer to the script will download an available YOLOv5 ONNX model and store it in the current model directory. Refer to the
Line 114: Line 114:
 ``` ```
  
-==== Model Conversion ====+==== Model conversion ====
 Switch to the ''rknn_model_zoo/examples/yolov5/python'' directory and run the ''convert.py'' script. This script converts Switch to the ''rknn_model_zoo/examples/yolov5/python'' directory and run the ''convert.py'' script. This script converts
 the original ONNX model to the RKNN model. Refer to the following commands: the original ONNX model to the RKNN model. Refer to the following commands:
Line 123: Line 123:
 ``` ```
  
-==== Run RKNN Python Demo ====+==== Run RKNN Python demo ====
 === Android === === Android ===
 Switch to the ''rknn_model_zoo/examples/yolov5/python'' directory, execute the ''yolov5.py'' script, and you can run the Switch to the ''rknn_model_zoo/examples/yolov5/python'' directory, execute the ''yolov5.py'' script, and you can run the
Line 135: Line 135:
 === Linux === === Linux ===
  
-Running on a Ubuntu 24.04 system.+Running on a Ubuntu 24.04 system which ship the python version of 3.12.
  
 Download the model zoo to your Edge 2L: Download the model zoo to your Edge 2L:
Line 147: Line 147:
 ```shell ```shell
 sudo apt update sudo apt update
-sudo apt install python3-venv -y+sudo apt install python3.12-venv -y
 ``` ```
  
Line 181: Line 181:
 ``` ```
  
-Copy the converted model to your Edge 2L:+Copy the converted model in [[#model-conversion | Model conversion]] to your Edge 2L:
  
 ```shell ```shell
 adb push examples/yolov5/model/yolov5s_relu.rknn /home/khadas/rknn_model_zoo/examples/yolov5/model adb push examples/yolov5/model/yolov5s_relu.rknn /home/khadas/rknn_model_zoo/examples/yolov5/model
 +```
 +
 +Modify the demo code:
 +
 +```diff
 +diff --git a/py_utils/rknn_executor.py b/py_utils/rknn_executor.py
 +index 21b4132..46b24d4 100644
 +--- a/py_utils/rknn_executor.py
 ++++ b/py_utils/rknn_executor.py
 +@@ -1,5 +1,5 @@
 +-from rknn.api import RKNN
 +-
 ++from rknnlite.api import RKNNLite as RKNN
 ++import numpy as np
 + 
 + class RKNN_model_container():
 +     def __init__(self, model_path, target=None, device_id=None) -> None:
 +@@ -9,10 +9,7 @@ class RKNN_model_container():
 +         rknn.load_rknn(model_path)
 + 
 +         print('--> Init runtime environment')
 +-        if target==None:
 +-            ret = rknn.init_runtime()
 +-        else:
 +-            ret = rknn.init_runtime(target=target, device_id=device_id)
 ++        ret = rknn.init_runtime()
 +         if ret != 0:
 +             print('Init runtime environment failed')
 +             exit(ret)
 +@@ -33,10 +30,12 @@ class RKNN_model_container():
 +         else:
 +             inputs = [inputs]
 + 
 ++        inputs = [np.expand_dims(item, axis=0) if len(item.shape) == 3 else item for item in inputs]
 ++
 +         result = self.rknn.inference(inputs=inputs)
 +     
 +         return result
 + 
 +     def release(self):
 +         self.rknn.release()
 +-        self.rknn = None
 +\ No newline at end of file
 ++        self.rknn = None
 +(myenv312) khadas@Khadas:~/rknn_model_zoo$ 
 ``` ```
  
Line 196: Line 241:
 {{:products:sbc:edge-2l:npu:yolov5-linux-demo.webp?600|}} {{:products:sbc:edge-2l:npu:yolov5-linux-demo.webp?600|}}
  
 +Deactivate the python virtual enviranment:
  
 +```shell
 +(myenv312) khadas@Khadas:~/rknn_model_zoo$ deactivate 
 +```
  
  
-==== Run RKNN C Demo ==== +==== Run RKNN C demo ==== 
-To run a RKNN C Demo, you need to first compile the C/C++ source code into an executable file. After that, push+To run a RKNN C demo, you need to first compile the C/C++ source code into an executable file. After that, push
 the executable file, model files, input images, and other related files to the development board. Finally, execute the executable file, model files, input images, and other related files to the development board. Finally, execute
 the executable file on the development board. the executable file on the development board.
Line 217: Line 266:
 ``` ```
  
-Push Files to the Edge 2L Board:+Push Files to the Edge 2L board:
 ```shell ```shell
 $ cd Projects/rknn_model_zoo $ cd Projects/rknn_model_zoo
Line 224: Line 273:
 ``` ```
  
-Run the Demo on Edge 2L Board:+Run the Demo on Edge 2L board:
 ```shell ```shell
 $ adb shell $ adb shell
Line 244: Line 293:
 ``` ```
  
-Push Files to the Edge 2L Board:+Push Files to the Edge 2L board:
 ```shell ```shell
 $ cd Projects/rknn_model_zoo $ cd Projects/rknn_model_zoo
Line 250: Line 299:
 ``` ```
  
-Run the Demo on Edge 2L Board:+Run the Demo on Edge 2L board:
 ```shell ```shell
 $ adb shell $ adb shell
Line 258: Line 307:
 ``` ```
  
- +==== View results ====
- +
-==== View Results ====+
 By default, the output image is saved at the path ''rknn_yolov5_demo/out.png''. You can use the adb tool to pull it By default, the output image is saved at the path ''rknn_yolov5_demo/out.png''. You can use the adb tool to pull it
 from the board to the local machine. In the local computer terminal, execute the following command: from the board to the local machine. In the local computer terminal, execute the following command:
Line 269: Line 316:
  
  
- +===== See also =====
-===== See Also =====+
  
 [[https://github.com/airockchip/rknn-toolkit2/tree/master/doc | RKNN Toolkit2 Documentations]] [[https://github.com/airockchip/rknn-toolkit2/tree/master/doc | RKNN Toolkit2 Documentations]]
  
Last modified: 2026/05/31 22:55 by nick

Page Tools

Khadas Site

Khadas Community

Khadas GitHub

Khadas Downloads