In scikit-learn, you can use the predict_proba method of a trained logistic regression model to get the probabilities of each class. Here is an example of how you can use it:
from sklearn.linear_model import LogisticRegression
from sklearn.model_selection import train_test_split
# Load the data and split it into training and test sets
X, y = load_data()
X_train, X_test, y_train, y_test = train_test_split(X, y, test_size=0.2, random_state=42)
# Train the logistic regression model
model = LogisticRegression()
model.fit(X_train, y_train)
# Get the predicted probabilities of the test set
probs = model.predict_proba(X_test)
# The probs array will have shape (n_samples, n_classes)
# and contain the probability of each sample belonging to each class.
# For example, the probability of the first sample belonging to class 0 is probs[0, 0]
# and the probability of the first sample belonging to class 1 is probs[0, 1]You can also use the predict method to get the predicted class labels, if you just want the classifications and not the probabilities.
# Get the predicted class labels of the test set
predictions = model.predict(X_test)
# The predictions array will have shape (n_samples,) and contain the predicted class labels
# For example, the predicted class label of the first sample is predictions[0]Keep in mind that the predict_proba method is only available for logistic regression models that are trained with the "ovr" (one-versus-rest) or "multinomial" multi-class strategies. If the model was trained with the "binary" strategy, it will only have two classes and predict_proba will only return the probability of the positive class.
