Model 100% trained in Google Colab but Lagging in

ShadesofPocket commented 2 months ago

Search before asking

[X] I have searched the HUB issues and found no similar bug report.

HUB Component

Models

Bug

As explained in the title, the model is 100% trained in googlecolab but Ultralytics HUB states training in progress 2% despite graphs for all Epochs being displayed

Environment

GoogleColab

Minimal Reproducible Example

No response

Additional

No response

UltralyticsAssistant commented 2 months ago

👋 Hello @ShadesofPocket, thank you for raising an issue about the Ultralytics HUB 🚀! We're here to help. Please check out our HUB Docs for more information:

Quickstart. Begin training and deploying YOLO models with ease.
Models: Training and Exporting. Find details on training YOLOv5 and YOLOv8 models and exporting them in various formats.
Integrations. Explore integration options, including TensorFlow and OpenVINO.

If this is a 🐛 Bug Report, please provide a minimum reproducible example (MRE) along with any screenshots to help us troubleshoot the issue effectively. You can learn more about creating an MRE here.

If this is a ❓ Question, please include details about your dataset, model, and environment to help us assist you better.

An Ultralytics engineer will review your issue soon. Thanks for your patience! 😊

sergiuwaxmann commented 1 month ago

@ShadesofPocket It looks like your model's metrics and weights were not uploaded during training with Colab. Unfortunately, you'd have to start the training again as there isn't a way to fix this issue.

ShadesofPocket commented 1 month ago

@sergiuwaxmann,

Thank you for the response! Yes I restarted the training and seemed to work second time around.

Many thanks for your help!

sergiuwaxmann commented 1 month ago

@ShadesofPocket No worries! Apologies for the inconvenience.

ultralytics / hub