Learn Before
Managing new categories during a computer vision error analysis.
Case context: You are building a cat classifier and examining misclassified images. You initially create categories for 'dogs' and 'blurry images'. Halfway through your dev set, you notice several misclassifications are actually lions.
Question: According to the iterative process described in Machine Learning Yearning, how should you handle the discovery of the 'lions' category, and what must you do with the images you already reviewed?
Sample answer: You should add 'lions' as a new error category. Because error analysis is an iterative process, you must then re-examine the misclassified images you have already categorized to see if any of them should be re-assigned to the new 'lions' category.
Key points:
- Add the newly discovered category.
- Error analysis is iterative.
- Re-examine previously categorized examples in light of the new category.
Rubric: Full credit if the learner correctly identifies that the new category should be added and explicitly states that previously examined images must be re-examined.
0
1
Tags
Machine Learning
Deep Learning
Supervised Learning
Dive into Deep Learning @ D2L
Data Science
Machine Learning Strategy
Machine Learning Yearning @ DeepLearning.AI
Related
Which of the following best describes the nature of error analysis in machine learning, according to Machine Learning Yearning?
True or False: According to Machine Learning Yearning, you must define all error categories before you begin reviewing misclassified examples.
In Machine Learning Yearning, Ng states that after looking at a couple of misclassified images, you might come up with a few ideas for error _____.
Match each phase of the iterative error analysis process to its correct description.
Place the steps of one iterative error analysis cycle in the correct order as described in Machine Learning Yearning.
During error analysis you discover a new error category after completing your first manual categorization pass. What should you do next?
True or False: In Machine Learning Yearning, Ng indicates that discovering new error categories during analysis may require revisiting previously examined examples.
After thinking of new categories during error analysis, you should re-examine examples in light of the _____ categories.
Match each characteristic of the error analysis process to the label that best describes it, as presented in Machine Learning Yearning.
Order the reasoning steps that explain why error analysis benefits from being iterative rather than a single pass.
Explain the iterative nature of error analysis.
Managing new categories during a computer vision error analysis.
Starting error analysis without predefined categories.