Skip to content

Add robust dataset loader for inconsistent .npy formats (fixes #178)#213

Open
kamalahasiniburra wants to merge 1 commit intoML4SCI:mainfrom
kamalahasiniburra:fix-robust-dataset-loader
Open

Add robust dataset loader for inconsistent .npy formats (fixes #178)#213
kamalahasiniburra wants to merge 1 commit intoML4SCI:mainfrom
kamalahasiniburra:fix-robust-dataset-loader

Conversation

@kamalahasiniburra
Copy link
Copy Markdown

Summary

Addresses Issue #178. This PR adds a robust dataset loader that handles inconsistent .npy file formats found across DeepLense sub-projects.

Features

  • DatasetLoader class with safe .npy loading (handles corrupted files gracefully)
    • Automatic shape validation and standardization (grayscale, channels-first/last)
    • Auto-detection of normalization ranges (uint8, [0,1], [-1,1])
    • Nearest-neighbor resize without external dependencies
    • Dataset info summary utility

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.

Add robust dataset loader utility for inconsistent .npy formats

1 participant