Skip to content

Conversation

@Shrishagk
Copy link

Summary

This PR improves usability, reproducibility, and robustness of the Super Resolution module by addressing dataset setup issues and a data loading bug.


Changes Made

1️⃣ Dataset Generation Script Added

• Introduced generate_dataset.py
• Automatically generates HR–LR image pairs
• Removes need for manual preprocessing
• Prevents FileNotFoundError during training


2️⃣ README Improved

• Added clear dataset setup instructions
• Documented directory structure
• Explained how to run dataset generation
• Reduced onboarding friction for new contributors


Impact

✔ Module can now be set up without hidden steps
✔ Pipeline becomes reproducible
✔ New contributors can run experiments easily
✔ Prevents runtime crashes


Testing

• Verified dataset generation script runs successfully
• Confirmed no FileNotFoundError after dataset setup
• Tested get_data_diff.py on small datasets without errors


Type of Contribution

  • Bug fix
  • Documentation improvement
  • New feature (utility script)

Closes: #135

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

Dataset setup is unclear and causes errors when running Super Resolution pipeline

1 participant