Assigning class weights during XGBoost training phase

Question

I am using the following code to train a XGBoost model:

# Calculate class weights for cost-sensitive learning
class_weights = compute_sample_weight(class_weight='balanced', y=y_train)

# Define a parameter distribution for randomized search
param_dist = {
    'eta': uniform(0.01, 0.3),
    'max_depth': randint(3, 10),
    'min_child_weight': randint(1, 10),
    'subsample': uniform(0.5, 1),  # Adjust range for subsample
    'colsample_bytree': uniform(0.5, 1),  # Adjust range for colsample_bytree
    'n_estimators': randint(50, 1000),  # Change the range to 1-1000
    'objective': ['binary:logistic'],
    'eval_metric': ['auc']
}

# Create an XGBoost classifier
xgb_classifier = xgb.XGBClassifier()

# Perform randomized search with cross-validation to find optimal parameters
random_search = RandomizedSearchCV(estimator=xgb_classifier,
                                   param_distributions=param_dist, n_iter=60, cv=5,
                                   scoring='roc_auc', n_jobs=-1, random_state=10)
random_search.fit(X_train, y_train, sample_weight=class_weights)

# Get the best parameters from the randomized search
best_params = random_search.best_params_

print("Best Parameters:", best_params)

# Use the best parameters to fit the XGBoost model
best_xgb_model = xgb.XGBClassifier(**best_params)
best_xgb_model.fit(X_train, y_train)

I have specified the weights in random_search fit. My question is, does this adjust the weights during training fit or should I include "sample_weights=class_weights" in best_xgb_model.fit as an argument?

Please help! Thanks !