the models available on Ultralytics' hub?

[jamesparkskku]

the models available on Ultralytics' hub?

[pderrenger]

@jamesparkskku hello!

Thank you for reaching out and for checking the existing issues and discussions before posting your question. It's great to see your interest in customizing and training YOLO models using Ultralytics HUB.

To address your query, yes, you can modify the YOLOv10 network and train it using a custom YAML file. Here’s a step-by-step guide to help you get started:

1. Upload Your Dataset: First, ensure your dataset is uploaded to the Ultralytics HUB. You can do this directly from the Home page by clicking on "Upload Dataset." For more details, you can refer to the datasets documentation.

2. Create a Project: Next, create a new project by selecting "Create Project" on the Home page. This will help you organize your work and keep track of your models and datasets.

3. Modify the YAML File: Customize your YAML file according to your project requirements. This file should define your dataset paths, number of classes, and other relevant parameters.

4. Train Your Model: Once your dataset and YAML file are ready, you can start training your model. On the Home page, select "Train Model" and choose the appropriate settings, including your custom YAML file. The Ultralytics HUB will handle the training process for you.

Here’s a basic example of what your YAML file might look like:

train: /path/to/your/train/images
val: /path/to/your/val/images

nc: 80  # number of classes
names: ['class1', 'class2', 'class3', ...]  # list of class names

If you encounter any capacity issues on your local machine or Colab, the Ultralytics HUB provides a robust cloud-based solution that can handle larger datasets and more intensive training processes.

Please ensure you are using the latest versions of the Ultralytics packages to avoid any compatibility issues. If the problem persists, feel free to provide more details, and we’ll be happy to assist you further.

For additional resources and detailed instructions, you can visit our quickstart guide.

Happy training! 🚀