Drift and model fit · Question 40 of 60
Which values indicate underfitting?
1 answer is correct. Medium
Show answer
Answer
- A Consistently low metrics such as accuracy or F1-score on training and validation data
Explanation
If accuracy, precision, recall or F1-score are persistently low on the training data and likewise on the validation data, the model is underfitted. If, on the other hand, it is markedly weaker on the test data than on the validation data, that points to overfitting.