Search⌘ K
Home
Courses
Business Machine Learning

Solution: Find the Best Combination of Parameters

Explore how to find the best combination of parameters for logistic regression models using grid search. Understand the process of scaling data, splitting datasets, training models, and evaluating performance to enhance predictive accuracy in classification tasks.

We'll cover the following...

Task

The task here is to find the best combination of parameters for the logistic regression model using grid search.

Solution

The provided workspace contains the code solution for the task mentioned above.

Python 3.8
X=df.drop(columns="target") # features
y=df["target"] # target
# features scaling
scaler=StandardScaler() # scaler instance
X=scaler.fit_transform(X) # scaling features
# train-test data sets
X_train, X_test, y_train, y_test = train_test_split(X, y, random_state = 7135, stratify = y)
logR=LogisticRegression() # model instance with default parameters
logR.fit(X_train, y_train) # training or fitting
print("mean cross-validation score with default parameters:",cross_val_score(logR,X_train,y_train,cv=5).mean())
print("\n <<< Grid Search >>> \n")
grid_values = {'penalty': ['l1', 'l2'],'C':[1, 0.1, 10, 100, 1000]}
logR=LogisticRegression() # model instance
grid = GridSearchCV(logR, grid_values, cv=5, scoring='accuracy')
grid.fit(X_train, y_train)
print("best parameters:", grid.best_params_)
print("\nmean cross-validation score with optimal parameters:",cross_val_score(grid,X_train,y_train,cv=5).mean())
...