Jasmine is testing a machine learning (ML) algorithm that classifies images as being ducks or not ducks. Her model has been built and trained. The testing data consists of 10 images of ducks, and the model correctly identifies them (within an appropriate margin of error) to be ducks. What is wrong with her testing procedure?
Responses which is the true correct a b c or d and give me your best reasoning
The testing data is outdated.
The testing data is outdated. - no response given
The testing data is homogenous.
The testing data is homogenous. - not selected, this is the correct answer
The testing data is biased.
The testing data is biased. - incorrect
The testing data is too large.
The correct answer is: The testing data is homogenous.
The issue with Jasmine's testing procedure is that the testing data is too similar or homogenous. In this case, the testing data consists only of images of ducks, which means that the model has only been tested on one type of data. This can lead to overfitting, where the model performs well on the specific testing data but may not generalize well to new, unseen data. To properly evaluate the model's performance, Jasmine should include a variety of images that are not ducks in her testing data set.