How should training resolution be chosen for a CNN or ViT?
mediumAnswer
- For CNNs: bigger resolution → higher accuracy up to a point (diminishing returns and OOM).
- Common choices: 224 for ResNet baselines, 256/288/380/456 for EfficientNet variants (compound scaling), 384/512 for ViT-Large.
- Train at moderate resolution, then fine-tune at higher resolution ('progressive resizing') for a cheap accuracy boost.
- For detection / segmentation, resolution matters more — small objects need high input res.
Check yourself — multiple choice
- Always 224
- Bigger up to diminishing returns; train mid-res then fine-tune at higher res (progressive resizing) for cheap gains
- Only 1024×1024
- Resolution never matters
Progressive resizing: train small then fine-tune at larger res → cheap accuracy boost.
#computer-vision#training
Practise Deep Learning
214 interview questions in this topic.
Related questions
- What is a solid default augmentation recipe for training a modern vision model?
- What are vanishing and exploding gradients, and how do you fix them?
- What does Batch Normalization do?
- How does dropout work and when is it applied?
- Why use a learning-rate schedule (warmup + cosine decay)?
- What inductive biases do CNNs have?