The two workhorses

• Regression — predict a number (house price, revenue, temperature).
• Classification — predict a category (spam/not-spam, churn/stay, disease/healthy).

The scikit-learn pattern (it's always the same)

from sklearn.linear_model import LogisticRegression
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)
model = LogisticRegression()
model.fit(X_train, y_train)          # learn
preds = model.predict(X_test)        # predict

Common algorithms to know

• Linear / Logistic Regression — simple, interpretable baselines.
• Decision Trees & Random Forests — flexible, strong defaults.
• Gradient Boosting (XGBoost) — often the top performer on tabular data.

Tip: always start with a simple baseline. If a linear model does well, you may not need anything fancier.