1.6 - Model Selection
🔖
One has to experiment on different models and check its performance with test data. The model with high accuracy will predict outcome better.
Use frameworks like TensorFlow, PyTorch, or Scikit-learn to build and experiment with different models.
Logistic Regression
It predicts the probability of an instance belonging to a particular class. In other words, it predicts which class a new data will belongs to based on outcome of probability.
This algorithm uses sigmoid function, which is a mathematical function used to map the predicted values to probabilities. It maps any real value into another value within a range of 0 and 1.
# Logistic regression
from sklearn.linear_model import LogisticRegression
lr = LogisticRegression()
model_lr = lr.fit(X_train, y_train)
# predict outcome with test data
y_pred_lr = model_lr.predict(X_test)
Then evaluate the model's performance using a test dataset.