Feature Importance and Model Explainability

Learn about the regression coefficients, the odds ratio of the coefficients, permutation feature importance, and model explainability.

Let's look at the linear models (linear regression, logistic regression, and their regularization extensions). We find a set of coefficients (β\beta values) for features in the training dataset, and then the weighted sum (coefficient_1×\timesfeature_1++ \cdots) helps us to make the predictions.

Regression coefficients

Technically, these coefficients (the β\beta values) can provide a basis for a crude way of scoring the respective feature's importance. This guides us to the usefulness of certain features for predicting the target values, given that the features are on the same scale. We can get the coefficients of our trained models and create a beautiful histogram. Let's get one for our model logR and explore how the model worked while training using (X_train_s, y_train).

Get hands-on with 1200+ tech skills courses.