This project is an image processing project that classifies objects in traffic with the Yolov8 model using the BDD100K dataset.
-The YOLO.ipynb file starts the training of the yolo model.
-reset.ipynb file deletes the "runs" folder after the yolo model is created.
-The controller.ipynb file classifies the image path you provide using the yolo model.
-Our Yolo model was trained for 10 epochs, but we didn't observe any overfitting in the models we experimented with later, which were around 50 epochs.
BDD100K dataset :"https://www.kaggle.com/datasets/awsaf49/bdd100k-dataset?select=bdd100k"
Don't forget to change the local paths in the file.
If you have problems with the project or files, please contact with me ("https://www.linkedin.com/in/yunusdermencioglu/")