PROFESSIONAL-DATA-ENGINEER · Question #327
You are developing a model to identify the factors that lead to sales conversions for your customers. You have completed processing your data. You want to continue through the model development lifecy
The correct answer is C. Delineate what data will be used for testing and what will be used for training the model.. After data processing, the next step in the model development lifecycle is to split data into training and test sets before training begins.
Question
Options
- AUse your model to run predictions on fresh customer input data.
- BMonitor your model performance, and make any adjustments needed.
- CDelineate what data will be used for testing and what will be used for training the model.
- DTest and evaluate your model on your curated data to determine how well the model performs.
How the community answered
(65 responses)- A8% (5)
- B2% (1)
- C86% (56)
- D5% (3)
Why each option
After data processing, the next step in the model development lifecycle is to split data into training and test sets before training begins.
Running predictions on fresh data is the final production inference step, which occurs only after training, evaluation, and deployment are complete.
Monitoring model performance occurs post-deployment, not before the model has been trained or evaluated.
Splitting data into training and testing subsets is a prerequisite to model training. Without a holdout test set defined upfront, there is no unbiased dataset available to evaluate model generalization after training, which is a foundational step in any supervised ML workflow.
Evaluating the model on curated data comes after the train/test split and after training, not before the data has been partitioned.
Concept tested: ML model development lifecycle: train/test data splitting
Source: https://developers.google.com/machine-learning/crash-course/training-and-test-sets/splitting-data
Topics
Community Discussion
No community discussion yet for this question.