Introduction to machine learning · Question 44 of 60
After tuning, a model achieves 94% accuracy on the validation data but only 78% on the independent test dataset. Which conclusions are appropriate?
2 answers are correct. Hard
Show answer
Answer
- C The model may not meet the criteria and may need to be generated again
- D If necessary, one has to go back to data preparation and train a new model
Explanation
The test results are compared with the evaluation results. If performance with independent test data is significantly lower, it may be necessary to return to model generation or even to the activity "prepare & test data". The workflow is explicitly iterative.