Welcome to RDK S600 Developer Kit repository

Model Training - Conversion

After completing the dataset annotation tutorial content, we can use the mainboard to start training the model.

This tutorial only covers CLI model training and conversion. You can refer to the official website to modify Python examples.
Due to the performance limitations of the RDK S600 mainboard, model training takes a long time, please be patient.

1、Model Training

Use CLI command to train the model directly: Copy the yolo26n.pt file to the configuration file directory, then open the terminal in the configuration file directory:

xxxxxxxxxx
cd /home/sunrise/ultralytics26/data/yahboom_data/orange_data

xxxxxxxxxx
yolo detect train data=orange.yaml model=yolo26n.pt epochs=100 imgsz=640

data: Dataset configuration file

model: Pre-trained model file

epochs: Number of training epochs

imgsz: Input specified image size

2、Model Conversion

The final model will be generated in the runs folder: Generally select the best.pt file for use

xxxxxxxxxx
/home/sunrise/ultralytics26/data/yahboom_data/orange_data/runs/detect/train/weights

Convert pt model to ncnn model:

xxxxxxxxxx
cd  /home/sunrise/ultralytics26/data/yahboom_data/orange_data/runs/detect/train/weights

xxxxxxxxxx
yolo export model=best.pt format=ncnn

References

Model Training with Ultralytics YOLO - Ultralytics YOLO Documentation