Khadas Docs

Amazing Khadas, always amazes you!

User Tools

Site Tools


products:sbc:edge2:npu:npu-convert

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:edge2:npu:npu-convert [2023/08/22 05:55]
louis
products:sbc:edge2:npu:npu-convert [2024/04/25 03:47] (current)
louis
Line 1: Line 1:
-Edge2 NPU Model Convert 修改稿+{{indexmenu_n>3}} 
 ====== NPU Model Convert ====== ====== NPU Model Convert ======
  
Line 25: Line 26:
 ===== Get Convert Tool ===== ===== Get Convert Tool =====
  
-Download Tool from [[https://github.com/rockchip-linux/rknn-toolkit2.git|Rockchip Github]].+Download Tool from [[gh>rockchip-linux/rknn-toolkit2]].
  
 ```shell ```shell
Line 44: Line 45:
 ===== Convert Model ===== ===== Convert Model =====
  
-Converting model has five main steps. Create RKNN object, pre-process config, load model, build model and export RKNN model. Here, take ''yolov5'' ‘’onnx’’ as an example.+Converting model has five main steps. Create RKNN object, pre-process config, load model, build model and export RKNN model. Here, take ''yolov5'' ''onnx'' as an example.
  
 Create RKNN object.  Create RKNN object. 
  
-```shell+```python
 # Create RKNN object # Create RKNN object
 rknn = RKNN(verbose=True)  rknn = RKNN(verbose=True) 
Line 55: Line 56:
 Pre-process config. Pre-process config.
  
-```shell+```python
 # pre-process config # pre-process config
 print('--> Config model') print('--> Config model')
Line 65: Line 66:
   * **std_values** - The variance of normalization parameter.   * **std_values** - The variance of normalization parameter.
 model input = (image – mean_values) / std_values model input = (image – mean_values) / std_values
-  * **target_platform** - Chooses ‘’rk3588’’.+  * **target_platform** - Chooses ''rk3588''.
  
 Load model. Load model.
  
-```shell+```python
 # Load ONNX model # Load ONNX model
 print('--> Loading model') print('--> Loading model')
-ret = rknn.load_onnx(model=./yolov5.onnx)+ret = rknn.load_onnx(model='./yolov5.onnx')
 if ret != 0: if ret != 0:
     print('Load model failed!')     print('Load model failed!')
Line 83: Line 84:
 Load other platform model. Load other platform model.
  
-```shell+```python
 # Load pytorch model # Load pytorch model
 print('--> Loading model') print('--> Loading model')
Line 130: Line 131:
 ``` ```
  
-  * **inputs/outputs** - Only use in ‘’tensorflow’’ model. It is the name of inputs/outputs.+  * **inputs/outputs** - Only use in tensorflow model. It is the name of inputs/outputs.
   * **input_size_list** - The size and channels of input.   * **input_size_list** - The size and channels of input.
  
 Build model Build model
  
-```shell+```python
 # Build model # Build model
 print('--> Building model') print('--> Building model')
-ret = rknn.build(do_quantization=True, dataset=./dataset.txt)+ret = rknn.build(do_quantization=True, dataset='./dataset.txt')
 if ret != 0: if ret != 0:
     print('Build model failed!')     print('Build model failed!')
Line 150: Line 151:
 Export RKNN model Export RKNN model
  
-```shell+```python
 # Export RKNN model # Export RKNN model
 print('--> Export rknn model') print('--> Export rknn model')
-ret = rknn.export_rknn(export_path=./yolov5_int8.rknn)+ret = rknn.export_rknn(export_path='./yolov5_int8.rknn')
 if ret != 0: if ret != 0:
     print('Export rknn model failed!')     print('Export rknn model failed!')
Line 162: Line 163:
   * **export_path** - The path of rknn model.   * **export_path** - The path of rknn model.
  
-All the above codes can be found in ‘’rknn-toolkit2/examples’’. There are all platforms we support now. Choose ‘’rk3588’’ in rknn-toolkit2/examples/onnx/yolov5/test.py and run the file to convert model.+All the above codes can be found in ''rknn-toolkit2/examples''. There are all platforms we support now. Choose ''rk3588'' in ''rknn-toolkit2/examples/onnx/yolov5/test.py'' and run the file to convert model.
  
 ```diff patch ```diff patch
Line 178: Line 179:
 ``` ```
  
-Run test.py to generate rknn model.+Run ''test.py'' to generate rknn model.
  
 ```shell ```shell
Line 190: Line 191:
 ===== See Also ===== ===== See Also =====
  
-For more usage, please refer to the related documents under ‘’doc’’.+For more usage, please refer to the related documents under ''doc''.
  
  
Last modified: 2023/08/22 05:55 by louis