Is this seemingly faster way of hyperparameter evolution going to be effective?

[NazmulTakbir]

❔Question

I want to perform hyperparameter evolution for my custom training data. Since it is a very time consuming process, can I use much smaller size images for hyperparameter evolution. And then use those evolved hyperparameters to train a network with much larger image sizes.

Additional context

I want to perform hyperparameter evolution mainly to solve the problem of very high class imbalance in the custom training data, as it has been suggested here - #1115 (comment)

[github-actions]

Hello @NazmulTakbir, thank you for your interest in our work! Please visit our Custom Training Tutorial to get started, and see our Jupyter Notebook , Docker Image, and Google Cloud Quickstart Guide for example environments.

If this is a bug report, please provide screenshots and minimum viable code to reproduce your issue, otherwise we can not help you.

If this is a custom model or data training question, please note Ultralytics does not provide free personal support. As a leader in vision ML and AI, we do offer professional consulting, from simple expert advice up to delivery of fully customized, end-to-end production solutions for our clients, such as:

• Cloud-based AI systems operating on hundreds of HD video streams in realtime.
• Edge AI integrated into custom iOS and Android apps for realtime 30 FPS video inference.
• Custom data training, hyperparameter evolution, and model exportation to any destination.

For more information please visit https://www.ultralytics.com.

[glenn-jocher]

@NazmulTakbir correlation between your base scenario and your hyperparameter evolution scenario is up to you. As the two diverge, the correlation in improvements between the two will also naturally diverge.