Randomly Trained XGBoost Model Generator

This function trains an XGBoost model for classification or regression. It randomly selects model parameters and trains the model using random sample training data.

Technology Stack : XGBoost, NumPy

Code Type : XGBoost model training function

Code Difficulty : Intermediate

                
                    
import xgboost as xgb
import numpy as np
import random

def random_xgb_model(X, y):
    # Define the parameters for the XGBoost model
    params = {
        'max_depth': random.choice([3, 5, 7, 9, 11]),
        'eta': random.uniform(0.01, 0.3),
        'subsample': random.uniform(0.5, 1.0),
        'colsample_bytree': random.uniform(0.5, 1.0),
        'objective': 'binary:logistic' if random.choice([True, False]) else 'reg:squarederror'
    }
    
    # Create the XGBoost DMatrix
    dtrain = xgb.DMatrix(X, label=y)
    
    # Train the model
    model = xgb.train(params, dtrain, num_boost_round=10)
    
    return model