Model Debugger Quiz
5 questions Pass: 70% +25 pts (50 with Pro)
Quiz covering Model Analysis
What you'll be asked
- Which of the following metrics is most useful for identifying if a classification model is failing on a specific minority class?
- You notice your model performs exceptionally well on the training set but poorly on the validation set. What is the most likely cause?
- When performing error analysis, you discover the model consistently misclassifies images of cats as dogs only in low-lighting conditions. What should be your next step?
- If a model's feature importance scores (e.g., via SHAP or Permutation Importance) reveal that a target-related variable is the strongest predictor by a massive margin, what should you immediately investigate?
- During model debugging, you use Partial Dependence Plots (PDPs). What specific insight do these plots provide that a global feature importance metric does not?
Answers and explanations come after you submit.
Course Content
- Data Cleaning Techniques
- Data Cleaning Techniques Quiz5q
- Feature Scaling Normalization
- Feature Scaling Normalization Quiz5q
- Encoding Techniques
- Encoding Techniques Quiz5q
- Glue DataBrew
- Glue DataBrew Quiz5q
- SageMaker Feature Store
- SageMaker Feature Store Quiz5q
- Ground Truth Labeling
- Ground Truth Labeling Quiz5q
- SageMaker Training Jobs
- SageMaker Training Jobs Quiz5q
- Hyperparameter Tuning
- Hyperparameter Tuning Quiz5q
- Distributed Training
- Distributed Training Quiz5q
- Fine-Tuning Models
- Fine-Tuning Models Quiz5q
- Regularization Techniques
- Regularization Techniques Quiz5q
- Model Registry
- Model Registry Quiz5q
- Ensemble Methods
- Ensemble Methods Quiz5q