YOLOv8-16bit/docs/sdk.md

Using YOLO models

This is the simplest way of simply using yolo models in a python environment. It can be imported from the ultralytics module.

!!! example "Usage" === "Training" ```python from ultralytics import YOLO

    model = YOLO("yolov8n.yaml")
    model(img_tensor) # Or model.forward(). inference.
    model.train(data="coco128.yaml", epochs=5)
    ```

=== "Training pretrained"
    ```python
    from ultralytics import YOLO

    model = YOLO("yolov8n.pt") # pass any model type
    model(...) # inference
    model.train(epochs=5)
    ```

=== "Resume Training"
    ```python
    from ultralytics import YOLO

    model = YOLO()
    model.resume(task="detect") # resume last detection training
    model.resume(model="last.pt") # resume from a given model/run
    ```

=== "Visualize/save Predictions"
```python
from ultralytics import YOLO

model = YOLO("model.pt")
model.predict(source="0") # accepts all formats - img/folder/vid.*(mp4/format). 0 for webcam
model.predict(source="folder", view_img=True) # Display preds. Accepts all yolo predict arguments

```

!!! note "Export and Deployment"

=== "Export, Fuse & info" 
    ```python
    from ultralytics import YOLO

    model = YOLO("model.pt")
    model.fuse()  
    model.info(verbose=True)  # Print model information
    model.export(format=)  # TODO: 

    ```
=== "Deployment"


More functionality coming soon

To know more about using YOLO models, refer Model class Reference

Model reference{ .md-button .md-button--primary}

---

Using Trainers

YOLO model class is a high-level wrapper on the Trainer classes. Each YOLO task has its own trainer that inherits from BaseTrainer. !!! tip "Detection Trainer Example" ```python from ultralytics.yolo import v8 import DetectionTrainer, DetectionValidator, DetectionPredictor

    # trainer
    trainer = DetectionTrainer(overrides={})
    trainer.train()
    trained_model = trainer.best

    # Validator
    val = DetectionValidator(args=...)
    val(model=trained_model)

    # predictor
    pred = DetectionPredictor(overrides={})
    pred(source=SOURCE, model=trained_model)

    # resume from last weight
    overrides["resume"] = trainer.last
    trainer = detect.DetectionTrainer(overrides=overrides)

    ```

You can easily customize Trainers to support custom tasks or explore R&D ideas. Learn more about Customizing Trainers, Validators and Predictors to suit your project needs in the Customization Section.

Customization tutorials{ .md-button .md-button--primary}