RandomSearch with XGBClassifier

randomsearch.py

import scipy.stats as st
from sklearn.model_selection import RandomizedSearchCV
from xgboost import XGBClassifier

# Preconfigure estimator and parameters
estimator = XGBClassifier(nthreads=-1)
params = {  
            "n_estimators": st.randint(3, 40),
            "max_depth": st.randint(3, 40),
            "learning_rate": st.uniform(0.05, 0.4),
            "colsample_bytree": st.beta(10, 1),
            "subsample": st.beta(10, 1),
            "gamma": st.uniform(0, 10),
            'objective': ['binary:logistic'],
            'scale_pos_weight': st.randint(0, 2),
            "min_child_weight": st.expon(0, 50),
        }

# Random Search Training with 5 folds Cross Validation
clf = RandomizedSearchCV(estimator, params, cv=5,
                         n_jobs=1, n_iter=100) 
                              
clf.fit(X_train, y_train)  
best_params = clf.best_params_
best_score = clf.best_score_

# Predict label from Test data
y_pred = clf.predict(X_test)