Lasso Regression (aolearn.lasso)

fit_lasso(X, y, lambda, learningRate, numIterations)

Trains a Lasso regression model (employing L1 regularization) using gradient descent.

Parameters:

• X (table): Training features.
• y (table): Training labels.
• lambda (number): Regularization strength.
• learningRate (number): Learning rate.
• numIterations (number): Number of iterations.

Returns:

• table: Learned coefficients (theta), including the bias term.

---

predict_lasso(theta, features)

Generates predictions using a fitted Lasso regression model.

Parameters:

• theta (table): Learned coefficients (including the bias) from fit_lasso.
• features (table): Features for a single data point.

Returns:

• number: Predicted value.