Purpose

• Evaluation metrics provide a numerical representation of ow well a model is doing.
• Metrics enable the comparison of different models or different versions of the same model.

Receiver Operating Characteristic Area Under the Curve (ROC AUC)

A metric that summarizes how well a classifier can distinguish between positive and negative classes

ROC AUC score is a single number that summarizes the classifier's performance across all possible classification thresholds. To get the score, you must measure the area under the ROC curve.

Root Mean Squared Error (RMSE)

<aside> 💡

Evaluation metrics for regression problems

</aside>

Measures the average magnitude of errors between predicted and actual values.

Advantages: Puts more weight on larger errors due to squaring.

Limitations: Can be influenced by extreme values, might not be robust to outliers

R-squared (R^2)

<aside> 💡

Evaluation metrics for regression problems

</aside>

Quantifies the proportion of the variance in the dependent variable that is predictable from the independent variables in a model.

Advantages: Easy to interpret. Indicates goodness of fit. ****

Limitations: Can be biased by the number of predictors.