Quiz · 8 questions
📊 Data Handling
Datasets, transforms, custom collate, HuggingFace, imbalanced data, and the GPU-feeding pipeline.
Level 0Tensor Curious
0 XP0/62 lessons0/13 achievements
0/120 XP to next level120 XP to go0% complete
Quiz
01What is the recommended transforms API in modern torchvision?
02Why should you NEVER apply random augmentation to validation data?
03What does
WeightedRandomSampler do?04Which DataLoader setting most directly accelerates CPU→GPU transfer?
05How do you split work across workers in an IterableDataset?
06What's the standard ImageNet normalization (mean, std)?
Hint
The numbers are unmemorable on purpose — but the second option starts 0.485...
07What do MixUp and CutMix have in common?
08Why is HuggingFace's datasets library memory-efficient on huge datasets?
Comments 0
🔔 Reply notifications (sign in)Sign in — Please sign in to comment.
No comments yet — be the first.