Regularizing models with Twitter data

You will work with the Twitter data expressing customers' sentiment about airline companies. The X matrix of features and y vector of labels have been created for you. In addition, the training and testing split has been performed. You can work with the X_train, X_test, y_train and y_test arrays directly.

You will train regularized and a more flexible models and evaluate them using different model performance metrics.

All required packages have been imported for you.

Cet exercice fait partie du cours

Sentiment Analysis in Python

Afficher le cours

Instructions

Train two logistic regressions: one with regularization parameter of 100 and a second of 0.1.
Print the accuracy scores of both models.
Print the confusion matrix of each model.

Exercice interactif pratique

Essayez cet exercice en complétant cet exemple de code.

# Build a logistic regression with regularizarion parameter of 100
log_reg1 = ____.____
# Build a logistic regression with regularizarion parameter of 0.1
log_reg2 = ____.____

# Predict the labels for each model
y_predict1 = log_reg1.predict(X_test)
y_predict2 = log_reg2.predict(X_test)

# Print performance metrics for each model
print('Accuracy of model 1: ', ____(____, ____))
print('Accuracy of model 2: ', ____(___, ____))
print('Confusion matrix of model 1: \n' , ____(____, ____)/len(y_test))
print('Confusion matrix of model 2: \n', ____(____, ____)/len(y_test))

Modifier et exécuter le code