Potential problems
Which of the following statements are TRUE regarding potential problems with holdout samples:
- A: Using different data splitting methods may lead to varying data in the final holdout samples.
- B: If you have limited data, your holdout accuracy may be misleading.
- C: There are no problems. Creating a single train and test sample is the only way to validate models.
- D: You shouldn't use holdout samples with limited data because you are limiting the potential training data.
This exercise is part of the course
Model Validation in Python
Hands-on interactive exercise
Turn theory into action with one of our interactive exercises
