Modelling

A confusion matrix plots predicted versus actual classes, allowing you to visualize true positives, false positives, true negatives, and false negatives to evaluate classifier performance. Regression residuals are used for continuous target variables, not classification, and hyperparameters are configuration settings rather than evaluation metrics.

BigQuery ML allows developers to build and execute machine learning models directly within BigQuery using standard SQL queries, which speeds up development by avoiding data movement. Option B is incorrect because BigQuery ML supports training complex models, including deep neural networks, directly in the cloud rather than locally.

A confusion matrix plots predicted versus actual classes, allowing you to visualize true positives, false positives, true negatives, and false negatives to evaluate classifier performance. A diagram for this topic isn't available yet — the worked example below walks the same reasoning step by step.

• !Option B is incorrect because BigQuery ML supports training complex models, including deep neural networks, directly in the cloud rather than locally.
• !A baseline model serves as a simple, easily interpretable starting point to establish a performance benchmark for more complex architectures.

Answer each from memory — you'll see instantly whether you're right and why.

Run a focused 10-question mini-mock on Modelling and see it stick.