Model porting best practices - Qualcomm Dragonwing Documentation

Export a custom YOLOv8 model using the QAIRT SDK

Prerequisites

Install the Qualcomm AI Runtime SDK on a host computer with Python >= 3.10 and PyTorch >=1.8. For more details, follow Install Qualcomm AI Runtime SDK. Run the following commands on the host computer.

Activate your virtual environment

source <venv_path>/bin/activate

Install the Ultralytics package and export the ONNX model

pip install ultralytics

yolo export model=yolov8s.pt imgsz=320 format=onnx opset=11 optimize=True simplify=true

Procedure

Convert the ONNX model to DLC

snpe-onnx-to-dlc -i yolov8s.onnx

Generate quantized DLC

Prepare the calibration data set

Gather 5-10 images that used during training and save these images in the input directory

Use the preprocess.py script to convert .jpg images into the RAW files required for quantization

In this example, the model uses an input dimension of 320x320.

Download the script

wget https://raw.githubusercontent.com/qualcomm/sample-apps-for-qualcomm-linux/refs/heads/main/qualcomm-linux/scripts/preprocess.py

Run the script with the following options

python preprocess.py <INPUT PATH> <OUTPUT PATH> 1 0

<INPUT PATH>: Folder containing the original images
<OUTPUT PATH>: Folder where the RAW files will be generated

Create an input.txt file containing the paths to all generated RAW files

The quantization process needs this file.

Quantize the model

Use snpe-dlc-quantize to convert the model to quantized DLC.

snpe-dlc-quantize --input_dlc yolov8s.dlc --input_list input.txt

Run the demo

Download the labels file

See Download model and label files.

On the host computer, set the user environment variable

export USER=ubuntu

Push the test video file to /etc/media on the device

scp <video file> $USER@<IP address of target device>:/home/ubuntu/
ssh ubuntu<ip-address>
cp /home/ubuntu/<video file> /etc/media/
exit

Push the quantized YoloV8 model to the device

scp <model file> $USER@<IP address of target device>:/home/ubuntu/
ssh ubuntu<ip-address>
cp /home/ubuntu/<model file> /etc/models/
exit

Retrieve the output tensor of the model, for example output0

Model output tensor name in Netron graph viewer

ssh $USER@<IP_ADDRESS_OF_TARGET_DEVICE>

In the new shell, run the following command

gst-launch-1.0 -e filesrc location=<video file path> ! qtdemux ! queue ! h264parse ! v4l2h264dec capture-io-mode=4 output-io-mode=4 ! video/x-raw,format=NV12 ! queue ! tee name=split split. ! queue ! qtivcomposer name=mixer ! queue ! waylandsink fullscreen=true sync=true split. ! queue ! qtimlvconverter ! queue ! qtimlsnpe delegate=dsp model=<model file path> tensors="<output0>" ! queue ! qtimlpostprocess settings="{\"confidence\": 51.0}" results=10 module=yolov8 labels=<label file path> ! video/x-raw,width=640,height=360 ! queue ! mixer.

Replace the following placeholders with the appropriate file paths:
<video file path>: Path to the input video file
<model file path>: Path to the YOLOv8 DLC model file
<label file path>: Path to the YOLOv8 label file

For example:

gst-launch-1.0 -e filesrc location=/etc/media/video.mp4 ! qtdemux ! queue ! h264parse ! v4l2h264dec capture-io-mode=4 output-io-mode=4 ! video/x-raw,format=NV12 ! queue ! tee name=split split. ! queue ! qtivcomposer name=mixer ! queue ! waylandsink fullscreen=true sync=true split. ! queue ! qtimlvconverter ! queue ! qtimlsnpe delegate=dsp model=/etc/models/yolov8s_quantized.dlc tensors="<output0>" ! queue ! qtimlpostprocess settings="{\"confidence\": 51.0}" results=10 module=yolov8 labels=/etc/labels/yolov8.json ! video/x-raw,width=640,height=360 ! queue ! mixer.

​Export a custom YOLOv8 model using the QAIRT SDK

​Prerequisites

​Procedure

​Run the demo

Export a custom YOLOv8 model using the QAIRT SDK

Prerequisites

Procedure

Run the demo