Image prediction

[smallMantou]

Search before asking

• I have searched the YOLOv8 issues and discussions and found no similar questions.

Question

Hello, I would like to ask how to use the weight file trained from my own dataset to predict multiple images in a folder.

Additional

No response

[Kayzwer]

@smallMantou model.predict("path/to/imgs")

[smallMantou]

Do I need to list each image for detecting multiple images?

[Kayzwer]

Do I need to list each image for detecting multiple images?

no just give the path to the img directory

[smallMantou]

Thank you for your answer.How do I know which model I am using when training my dataset, because I want to start training from scratch, but the yolo. yaml file contains n \ s \ m \ l.Grayscale_yolov5s.yaml is yolov8.yaml that I copied.

[glenn-jocher]

Hello! To determine which model configuration you are using when training from scratch, you should look at the YAML file you are specifying in your training command. Each model variant (n, s, m, l) has its own YAML file that defines the model's architecture and settings. If you've copied and renamed a YAML file, make sure to check its contents to see which model it's configured for.

For example, if you're using Grayscale_yolov5s.yaml and you've renamed it from a YOLOv8 model configuration, you should open that YAML file and review the parameters to confirm which model size it corresponds to (n, s, m, l). The model size is typically indicated by the depth and width parameters in the YAML file.

Here's a quick way to start training from scratch using a specific model configuration:

yolo train data=your_dataset.yaml model=your_model.yaml

Replace your_dataset.yaml with your dataset configuration file and your_model.yaml with the model configuration file you intend to use. This will ensure you are training with the correct model setup. Happy training! 🚀