How to train with new pictures?

[PengPeng-JunJun]

Search before asking

• I have searched the YOLOv5 issues and discussions and found no similar questions.

Question

Hi, I have to train as code YOLOV5-6.1 and get good result.Now , I add some new images to the dataset.I want to know how to continue training on the best.pt before.
And please tell me how to call this operation,thanks!

Additional

No response

[UltralyticsAssistant]

👋 Hello @PengPeng-JunJun, thank you for your interest in YOLOv5 🚀!

Please visit our ⭐️ Tutorials to get started with tasks like Custom Data Training and more advanced concepts.

For your question about continuing training, you can load your best.pt model and resume training with your updated dataset. This process is often referred to as "fine-tuning" or "continued training."

If this relates to a 🐛 Bug Report, please provide a minimum reproducible example to help us debug it.

If you have any custom training ❓ questions, please include details such as dataset examples and logs, and ensure you're following our Tips for Best Training Results.

Requirements

Ensure you have Python>=3.8.0 and install all dependencies in requirements.txt, including PyTorch>=1.8:

git clone https://github.com/ultralytics/yolov5
cd yolov5
pip install -r requirements.txt

Environments

YOLOv5 can be run in the following environments, which include all necessary dependencies:

• Notebooks with free GPU:
• Google Cloud: See GCP Quickstart Guide
• Amazon EC2: See AWS Quickstart Guide
• Docker: See Docker Quickstart Guide

Status

A green badge indicates all YOLOv5 GitHub Actions tests are passing.

Introducing YOLOv8 🚀

We are thrilled to introduce YOLOv8 🚀! Ideal for object detection, image segmentation, and classification tasks. Check out the YOLOv8 Docs and get started:

pip install ultralytics

An Ultralytics engineer will follow up with you soon! 😊