from sklearn.datasets import make_classification

# Define the dataset parameters
n_samples  = 20_000_000  # 10 million rows
n_features = 5           # 5 columns
n_classes  = 2           # Two classes

# Generate the dataset
X, y = make_classification(
    n_samples=n_samples,
    n_features=n_features,
    n_informative=n_features,
    n_classes=n_classes,
    n_redundant=0,
    n_repeated=0,
    random_state=42  
)