Khadas Docs

Amazing Khadas, always amazes you!

User Tools

Site Tools

Home / products / sbc / vim4 / npu / demos / yolov8n
products:sbc:vim4:npu:demos:yolov8n

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:vim4:npu:demos:yolov8n [2024/01/04 04:51]
louis 		products:sbc:vim4:npu:demos:yolov8n [2025/06/12 05:39] (current)
louis
Line 1: Line 1:
 ~~tag> NPU YOLO OpenCV VIM4 ~~ ~~tag> NPU YOLO OpenCV VIM4 ~~
 +
 +**Doc for version ddk-3.4.7.7**
  
 ====== YOLOv8n OpenCV VIM4 Demo - 2 ====== ====== YOLOv8n OpenCV VIM4 Demo - 2 ======
  
 {{indexmenu_n>2}} {{indexmenu_n>2}}
 +
 +===== Introduction =====
 +
 +YOLOv8n is an object detection model. It uses bounding boxes to precisely draw each object in image.
 +
 +Inference results on VIM4.
 +
 +{{:products:sbc:vim4:npu:demos:yolov8n-demo-output.webp?800|}}
 +
 +**Inference speed test**: USB camera about **150ms** per frame.
 +
 ===== Get Source Code ===== ===== Get Source Code =====
  
Line 12: Line 25:
 ``` ```
  
-Refer ''README.md'' to train a YOLOv8n model.+Refer ''README.md'' to train a YOLOv8n model. My version ''torch==1.10.1'' and ''ultralytics==8.0.86''.
  
 ===== Convert Model ===== ===== Convert Model =====
Line 20: Line 33:
 Follow Docker official documentation to install Docker: [[https://docs.docker.com/engine/install/ubuntu/|Install Docker Engine on Ubuntu]]. Follow Docker official documentation to install Docker: [[https://docs.docker.com/engine/install/ubuntu/|Install Docker Engine on Ubuntu]].
  
-Then fetch the prebuilt NPU Docker container and run it.+Follow the script below to get Docker image:
  
 ```shell ```shell
-$ docker pull yanwyb/npu:v1 +docker pull numbqq/npu-vim4
-$ docker run -it --name vim4-npu1 -v $(pwd):/home/khadas/npu \ +
- -v /etc/localtime:/etc/localtime:ro \ +
- -v /etc/timezone:/etc/timezone:ro \ +
- yanwyb/npu:v1+
 ``` ```
  
 ==== Get Model Conversion Tools ==== ==== Get Model Conversion Tools ====
  
-Get source [[gl>khadas/vim4_npu_sdk]].+Get source [[gh>khadas/vim4_npu_sdk]].
  
 ```shell ```shell
-$ git clone https://gitlab.com/khadas/vim4_npu_sdk+$ git lfs install 
 +$ git lfs clone https://github.com/khadas/vim4_npu_sdk 
 +$ cd vim4_npu_sdk 
 +$ ls 
 +adla-toolkit-binary  adla-toolkit-binary-3.1.7.4  convert-in-docker.sh  Dockerfile  docs  README.md
 ``` ```
 +
 +  * ''adla-toolkit-binary/docs'' - SDK documentations
 +  * ''adla-toolkit-binary/bin'' - SDK tools required for model conversion
 +  * ''adla-toolkit-binary/demo'' - Conversion examples
 +
 +<WRAP important>
 +If your kernel is older than 241129, please use branch npu-ddk-1.7.5.5.
 +</WRAP>
  
 ==== Convert ==== ==== Convert ====
Line 42: Line 63:
 After training model, modify ''ultralytics/ultralytics/nn/modules/head.py'' as follows. After training model, modify ''ultralytics/ultralytics/nn/modules/head.py'' as follows.
  
-```diff ultralytics/nn/modules/head.py.patch+```diff head.py
 diff --git a/ultralytics/nn/modules/head.py b/ultralytics/nn/modules/head.py diff --git a/ultralytics/nn/modules/head.py b/ultralytics/nn/modules/head.py
 index 0b02eb3..0a6e43a 100644 index 0b02eb3..0a6e43a 100644
Line 66: Line 87:
 +            dfl = self.cv2[i](x[i]).contiguous() +            dfl = self.cv2[i](x[i]).contiguous()
 +            cls = self.cv3[i](x[i]).contiguous() +            cls = self.cv3[i](x[i]).contiguous()
-+            results.append(torch.cat([cls, dfl], 1).permute(0, 2, 3, 1)) ++            results.append(torch.cat([cls, dfl], 1))
-+            # results.append(torch.cat([cls, dfl], 1))+
 +        return tuple(results) +        return tuple(results)
 + +
 ``` ```
 +
 +<WRAP important>
 +If you pip-installed ultralytics package, you should modify in package.
 +</WRAP>
  
 Create a python file written as follows to export ONNX model. Create a python file written as follows to export ONNX model.
Line 83: Line 107:
 $ python export.py $ python export.py
 ``` ```
 +
 +<WRAP important>
 +Use [[https://netron.app/ | Netron]] to check your model output like this. If not, please check your ''head.py''.
 +
 +{{:products:sbc:vim4:npu:demos:yolov8n-vim4-output-ddk-3-4-7-7.webp?600|}}
 +</WRAP>
  
 Enter ''vim4_npu_sdk/demo'' and modify ''convert_adla.sh'' as follows. Enter ''vim4_npu_sdk/demo'' and modify ''convert_adla.sh'' as follows.
Line 103: Line 133:
         --input-shapes  "3,640,640"  \         --input-shapes  "3,640,640"  \
         --dtypes "float32" \         --dtypes "float32" \
-        --inference-input-type float32 \ +        --quantize-dtype int16 --outdir onnx_output  \
- --inference-output-type float32 \ +
-        --quantize-dtype int8 --outdir onnx_output  \+
         --channel-mean-value "0,0,0,255"  \         --channel-mean-value "0,0,0,255"  \
 +        --inference-input-type "float32" \
 +        --inference-output-type "float32" \
         --source-file dataset.txt  \         --source-file dataset.txt  \
-        --disable-per-channel False \ 
-        --iterations 500 \ 
         --batch-size 1 --target-platform PRODUCT_PID0XA003         --batch-size 1 --target-platform PRODUCT_PID0XA003
 ``` ```
- 
-<WRAP important > 
-Please prepare about 500 pictures for quantification. If the pictures size is smaller than model input size, please resize pictures to input size before quantification. 
-</WRAP> 
  
 Run ''convert_adla.sh'' to generate VIM4 model. The converted model is ''xxx.adla'' in ''onnx_output''. Run ''convert_adla.sh'' to generate VIM4 model. The converted model is ''xxx.adla'' in ''onnx_output''.
Line 133: Line 157:
 ``` ```
  
-<WRAP important > +<WRAP important> 
-If your kernel version is 5.4 or earlier, please use tag ''ddk-1.7.5.5''. Tag ''ddk-2.3.6.7'' is for 5.15.+If your kernel is older than 241129, please use version before tag ddk-3.4.7.7.
 </WRAP> </WRAP>
  
Line 159: Line 183:
  
 # Run # Run
-$ sudo ./yolov8n -m ../data/yolov8n_int8.adla -p ../data/horses.jpg+$ ./yolov8n -m ../data/yolov8n_int8.adla -p ../data/horses.jpg
 ``` ```
  
Line 175: Line 199:
  
 # Run # Run
-$ sudo ./yolov8n_cap -m ../data/yolov8n_int8.adla -d 0 -w 1920 -h 1080+$ ./yolov8n_cap -m ../data/yolov8n_int8.adla -t usb -d 0
 ``` ```
  
Last modified: 2024/01/04 04:51 by louis

Page Tools

Khadas Site

Khadas Community

Khadas GitHub

Khadas Downloads