✅ Checkpoint: Computer Vision track
A mixed quiz on the whole Computer Vision track. Pass mark: 7 of 9 on the first try.
How this checkpoint works
Answer without looking back; only your first try counts.
Quiz
1. Which task fits "count every cell in a microscope image, with its location"?
2. A horizontally flipped training image has a box (x1, x2) = (10, 30) in an image 100 wide. The new box is…
3. Your classifier is over-confident on validation data. A cheap fix is…
4. Raising the NMS IoU threshold from 0.3 to 0.7 tends to…
5. Average precision for detection is…
6. Dice is 0.8. IoU is…
7. What do U-Net skip connections carry?
8. With 300 labelled specialist images, which usually beats CLIP zero-shot?
9. Before claiming a wound model works, the most important evaluation is…
📚 Further reading
Trusted, mostly free sources: official documentation, well-known textbooks and courses. When you want to confirm or go deeper on anything here, start with these.
- CS231n: Deep Learning for Computer Vision (Stanford) ↗ Notes on backprop, CNNs and training