Skip to content
ml-recall

#mean-accuracy

1 question from Segmentation

Timed test

1 question · 1 min · score and explanations at the end

Take a test
Q 1 / 1Score 0 / 0

Question 1: Short answer

mediumAI-checkedSegmentation

In a medical segmentation dataset, 95% of the pixels are background and 5% are lesion. A model predicts background for every pixel. Over the two classes (background and lesion), its pixel accuracy is , its mean (per-class) accuracy, the average of the fraction of each class's pixels predicted correctly, is , and its mean IoU is .

Fill in every blank (hints in the boxes). Enter moves to the next blank.