from pycaret.classification import *

# Import libraries
import numpy as np
import pandas as pd

# Set random seed for reproducibility
np.random.seed(42)

# Define number of samples
N = 30000

# Generate features
numeric_feature = np.random.normal(0, 1, N)
cat_feature = np.random.choice(31, N).astype(int)
cat_feature2 = np.random.choice(2, N).astype(str)

# Combine features into a single matrix
X = np.column_stack((cat_feature, cat_feature2))

# Generate target variable
y = np.random.binomial(1, 0.2, N)

# Convert to DataFrame and add target variable
df = pd.DataFrame(X, columns=['categorical-cardinal', 'categorical-binary'])
df['target'] = y

# Print the first five rows of the dataset
print("Synthetic Dataset for Machine Learning:\n")
print(df.head())

setup(df, target='target', numeric_features=[], categorical_features=['categorical-cardinal', 'categorical-binary'])

model = create_model('lightgbm')

plot_model(model, 'auc')

predict_model(model)