Image Classification Quiz

What you'll be asked

1. What is the primary goal of an image classification task in computer vision?
2. When training an image classification model, why is it important to normalize pixel values (e.g., scaling them to a range of 0 to 1)?
3. If a model achieves 99% accuracy on the training set but only 60% on the validation set, what is the most likely issue?
4. In the context of a Convolutional Neural Network (CNN), what is the function of the pooling layer?
5. You are building an image classifier for a dataset where 'cat' images make up 95% of the data and 'dog' images make up 5%. Which evaluation metric is most misleading in this scenario?

Answers and explanations come after you submit.