AI-900 Question #19: Real Exam Question with Answer & Explanation

Question

When training a model, why should you randomly split the rows into separate subsets?

Options

• Ato train the model twice to attain better accuracy
• Bto train multiple models simultaneously to attain better performance
• Cto test the model by using data that was not used to train the model

Explanation

Randomly splitting data into distinct training and testing subsets is essential to evaluate a model's performance on unseen data. This process ensures the model can generalize well and helps prevent overfitting.

Common mistakes.

• A. Training a model twice on the same data or different subsets without a proper test set does not inherently lead to better accuracy and can still result in an overfit model if not evaluated on unseen data.
• B. While multiple models can be trained, the primary reason for splitting data is not to train them simultaneously for better performance, but rather to properly evaluate a single model's generalization capability.

Concept tested. Data splitting for machine learning evaluation

Reference. https://learn.microsoft.com/en-us/azure/machine-learning/concept-train-test-validate-split?view=azureml-api-2

No community discussion yet for this question.