Reported in a candidate post, May 2025
Normalization, CNNs, pooling, segmentation, ROC-AUC versus PR-AUC and NaN loss
Filed under Meesho
The question
Explain layer versus batch normalization, convex versus non-convex losses, CNNs and pooling, segmentation basics, comparing ROC-AUC and PR-AUC under class imbalance, and causes of NaN loss; plus a discussion of an alternative solution for the candidate's hackathon project.
No write-up yet. Related questions below have one.
Related Meesho questions
Collected from public candidate posts and coding forums. Updated Nov 2026.