Feature Engineering for Modern Machine Learning with Scikit-LearnChapter 82
Answers
Section 1 of 2-~ 12 min read-Synced from Cuantum content
- B) Pipelines ensure transformations are consistently applied to both training and test data.
- B) To combine multiple transformations applied in parallel into a single dataset.
- B) A technique to select the most important features by recursively removing the least impactful features.
- C) When the dataset has a significant class imbalance.
- C) It generates synthetic samples by interpolating between existing minority samples.
- B) Time-Series Split Cross-Validation
- B) Accuracy does not account for model bias toward the majority class.
- B) F1 Score
- B) SMOTE can be applied in each cross-validation fold using a pipeline to balance classes in each fold.
- B) To ensure feature engineering steps are applied consistently across training and test data.