Question: 1. You train an image classification model which is meant to distinguish between images that contain a cat and those which don't. You are targeting

1. You train an image classification model which is meant to distinguish between images that contain a cat and those which don't. You are targeting a 2% error rate out-of- sample. Question (a): Assume your current model provides a training set error rate of 5% and a test set error rate of 6%. What would be a potential solution to improve your classification performance? Question (b): Provide one possible solution to improve your classification perfor- mance if your current model provides a training set error rate of 1% and a test set error rate of 4% instead
Step by Step Solution
There are 3 Steps involved in it
Get step-by-step solutions from verified subject matter experts
