Elastic Net Regression
#Elastic Net regression is a type of linear regression technique that combines
#the features of both L1 Lasso regression and L2 Ridge regression.
#Help prevent overfittting
#best used data set many features with some correlated
#The combined penalty term is controlled by two hyperparameters:alpha, L1 Ratio
import seaborn as sns import pandas as pd sns.get_dataset_names() 
tips = sns.load_dataset("tips") tips 
tips = pd.get_dummies(tips) X = tips.drop('tip', axis=1) y = tips['tip'] from sklearn.model_selection import train_test_split X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=19) from sklearn.preprocessing import StandardScaler scaler = StandardScaler() X_train = scaler.fit_transform(X_train) X_test = scaler.transform(X_test) from sklearn.linear_model import ElasticNet elastic_net = ElasticNet() elastic_net.fit(X_train, y_train) 
y_pred = elastic_net.predict(X_test) from sklearn.metrics import mean_absolute_error, mean_squared_error, r2_score mean_absolute_error(y_test, y_pred) 
mean_squared_error(y_test, y_pred) 
r2_score(y_test, y_pred) 
#l1_ratio: Another way to specify the balance between L1 and L2 regularization.
#Try different values of alpha
param_grid = { "alpha": [0.1, 0.3, 0.5, 0.7, 0.9], 'l1_ratio': [0.1, 0.3, 0.5, 0.7, 0.9],} from sklearn.model_selection import GridSearchCV elastic_cv = GridSearchCV(estimator=elastic_net, param_grid=param_grid, cv=3, scoring='neg_mean_squared_error', n_jobs=-1) elastic_cv.fit(X_train, y_train) 
y_pred = elastic_cv.predict(X_test) mean_absolute_error(y_test, y_pred) 
mean_squared_error(y_test, y_pred) 
r2_score(y_test, y_pred) 
