Visualize / Log Batch Size #966

Activity

1. WhoIsElMasri

   added

   enhancementNew feature or requestNew feature or request

   on Jan 1, 2025
2. UltralyticsAssistant

   added

   HUBUltralytics HUB issuesUltralytics HUB issues

   on Jan 1, 2025
3. UltralyticsAssistant commented on Jan 1, 2025

   

   UltralyticsAssistant

   on Jan 1, 2025

   Member

   More actions

   👋 Hello @WhoIsElMasri, thank you for raising an issue about Ultralytics HUB 🚀! It's great to see your interest in improving the platform. Your suggestion to visualize or log the batch size when using batch_size=-1 -> AUTO is quite valuable.

   If this is a ❓ Feature Request, further context or examples of how this would enhance your workflows might be helpful for our team. Feel free to provide any additional insights or specific applications.

   If this is a 🐛 Bug Report, please include any relevant details like steps to reproduce, environment setup, or screenshots, as this helps us investigate the matter more quickly.

   This is an automated response, but an Ultralytics engineer will review your issue and assist you as soon as possible. Thank you for your patience and contribution! 😊
4. pderrenger commented on Jan 2, 2025

   

   pderrenger

   on Jan 2, 2025

   Member

   More actions

   @WhoIsElMasri thank you for your suggestion! 😊

   Adding a feature to visualize or log the automatically determined batch size (batch_size=-1 -> AUTO) in the Ultralytics HUB interface is indeed a great idea. This could provide users with greater transparency and help them understand the batch size that is being computed based on their hardware capabilities.

   As you may already know, the batch_size=-1 setting leverages the autobatch function to compute the optimal batch size dynamically, ensuring efficient GPU memory utilization. For more details on how this works, you can refer to the autobatch documentation.

   We’ll take this feedback into account and consider it for future improvements to the HUB. In the meantime, if you’re looking to view the computed batch size during training, you can check the training logs, where this information should be displayed. Alternatively, if you're using Python scripts, you can explicitly call the autobatch function to compute and display the batch size:

   from ultralytics.utils.autobatch import autobatch
   
   optimal_batch_size = autobatch(model)  # Replace 'model' with your YOLO model instance
   print(f"Optimal Batch Size: {optimal_batch_size}")

   Feel free to share any additional thoughts or use cases for this feature. Your insights are always valued by the Ultralytics team and the YOLO community! 🚀
5. hamua-cmd commented on Jan 2, 2025

   

   hamua-cmd

   on Jan 2, 2025

   More actions

   Hey @pderrenger, thank you for the fast response.
   Where can I see the training logs in the HUB train interface? Did not find it yet
6. pderrenger commented on Jan 2, 2025

   

   pderrenger

   on Jan 2, 2025

   Member

   More actions

   Hi @hamua-cmd, you're very welcome! 😊

   In the Ultralytics HUB, training logs are accessible directly within the training interface. Once you start a training job, you can:

   1. Navigate to your project: Go to the "Projects" section where your training job is located.
   2. Select the specific run: Click on the training run you are interested in.
   3. View logs in real-time: You will see a live training dashboard that includes metrics like loss, mAP, and performance graphs, as well as text-based logs.

   The logs should display details such as the computed batch size (if batch_size=-1 was used), GPU utilization, and other training parameters. This helps ensure you can monitor your model's progress and configuration.

   If you are unable to locate these logs, could you confirm which step you’re on or share any additional details? I’ll be happy to guide you further. 🚀
7. github-actions commented on Nov 23, 2025

   

   github-actionsbot

   on Nov 23, 2025 – with GitHub Actions

   More actions

   👋 Hello there! We wanted to give you a friendly reminder that this issue has not had any recent activity and may be closed soon, but don't worry - you can always reopen it if needed. If you still have any questions or concerns, please feel free to let us know how we can help.

   For additional resources and information, please see the links below:

   • Docs: https://docs.ultralytics.com
   • HUB: https://hub.ultralytics.com
   • Community: https://community.ultralytics.com

   Feel free to inform us of any other issues you discover or feature requests that come to mind in the future. Pull Requests (PRs) are also always welcomed!

   Thank you for your contributions to YOLO 🚀 and Vision AI ⭐
8. github-actions

   added

   StaleStale and schedule for closing soonStale and schedule for closing soon

   on Nov 23, 2025
9. glenn-jocher commented on Jul 7, 2026

   

   glenn-jocher

   on Jul 7, 2026

   Member

   More actions

   Hi @WhoIsElMasri, closing the loop on this older request since HUB winds down on July 31, 2026: Ultralytics Platform covers this — each model's Overview tab lists the actual training arguments (epochs, batch size, image size, …), and the Train tab's Console subtab streams the full live training log, where the AutoBatch-computed batch size appears when you train with batch=-1. Docs: https://docs.ultralytics.com/platform/train/models/#overview-tab