integrations/tensorboard/

[giscus[bot]]

integrations/tensorboard/

Learn how to integrate YOLOv8 with TensorBoard for real-time visual insights into your model's training metrics, performance graphs, and debugging workflows.

https://docs.ultralytics.com/integrations/tensorboard/

[lauravchrobak]

It looks like the Usage code isn't complete. Looks like the first row is cut off?

[glenn-jocher]

@lauravchrobak hi there!

Thank you for pointing that out. It seems like the code snippet got truncated. Here's the complete usage example for integrating YOLOv8 with TensorBoard:

from ultralytics import YOLO

# Load a pre-trained model
model = YOLO('yolov8n.pt')

# Train the model
results = model.train(data='coco8.yaml', epochs=100, imgsz=640)

After running the training code, you can start TensorBoard to visualize the training metrics:

tensorboard --logdir path/to/your/tensorboard_logs

You can then access the TensorBoard dashboard at http://localhost:6006/ to monitor your training progress in real-time.

For more detailed guidance, you can visit our TensorBoard integration documentation.

If you encounter any issues, please ensure you are using the latest version of the Ultralytics package. If the problem persists, providing a minimum reproducible example would be very helpful. You can find more information on how to create one here.

Happy training! 🚀

[ruslanraupoff]

How to forward Tensorboard IP from Docker? Is it possible to use a yolo cli?

[glenn-jocher]

@ruslanraupoff to forward TensorBoard from Docker, map the port using -p 6006:6006 when running your container and start TensorBoard with tensorboard --logdir /path/to/runs. Yes, you can use YOLO CLI for training—logs are auto-saved to runs/ for TensorBoard visualization. See our Docker Quickstart for setup details.