Using Hyperparameter Tuning we want to optimize model performance. Please provide a summary for the code.filtered$\_$df$\_$model$[\text{'}$price$\_$scaled'$]=$ scaler$\_$price.fit$\_$transform$($filtered$\_$df$\_$model$[[\text{'}$pricfeatures $=$ filtered$\_$df$\_$model.drop$([\text{'}$price$\text{'},$ 'price$\_$scaled'$],$ axis$=1)$

target $=$ filtered$\_$df$\_$model$[\text{'}$price$\_$scaled'$]$X$\_$train, X$\_$test, y$\_$train, y$\_$test $=$ train$\_$test$\_$split$($features$,$ target, test$\_$size$=0\mathrm{.}15,$ rtrain$\_$data $=$ pd$.$concat$([$X$\_$train, y$\_$train$],$ axis$=1)$majority $=$ train$\_$data$[$train$\_$data$[\text{'}$price$\_$scaled'$]$ train$\_$data$[\text{'}$price$\_$scaled'$].$quantile$($

minority $=$ train$\_$data$[$train$\_$data$[\text{'}$price$\_$scaled'$]>=$ train$\_$data$[\text{'}$price$\_$scaled'$].$quantilemajority$\_$resampled $=$ resample$($majority$,$ n$\_$samples$=$len$($minority$)*2,$resampled$\_$train$\_$data $=$ pd$.$concat$([$majority$\_$resampled, minority$])$x$\_$train$\_$resampled $=$ resampled$\_$train$\_$data.drop$(\text{'}$price$\_$scaled', axis$=1)$

y$\_$train$\_$resampled $=$ resampled$\_$train$\_$data$[\text{'}$price$\_$scaled'$]$xgb$\_$model $=$ XGBRegressor$($random$\_$state$=42)$param$\_$grid $=\{\text{'}$learning$\_$rate': $[0\mathrm{.}01,0\mathrm{.}1,0\mathrm{.}2],$'subsample': $[0\mathrm{.}8,0\mathrm{.}9,1\mathrm{.}0],\}$grid$\_$search $=$ GridSearchCV$($estimator$=$xgb$\_$model, scoring$=$'neg$\_$mean$\_$squared$\_$error', n$\_$jobs$=-1,$Fit GridSearchCV

grid$\_$search.fit$($X$\_$train$\_$resampled, y$\_$train$\_$resampled$)$best$\_$params $=$ grid$\_$search.best$\_$params$\_$

best$\_$model $=$ grid$\_$search.best$\_$estimator$\_$

print$($f$"$Best Parameters: $\{$best$\_$params$\}")$xgb$\_$predictions $=$ best$\_$model.predict$($X$\_$test$)$from sklearn.metrics import mean$\_$absolute$\_$error, mean$\_$squared$\_$error, r$2\_$score

def evaluate$\_$model$($y$\_$true, y$\_$pred$)$:mse $=$ mean$\_$squared$\_$error$($y$\_$true, y$\_$pred$)$r$2=$ r$2\_$score$($y$\_$true, y$\_$pred$)$#Get metrics

xgb$\_$metrics $=$ evaluate$\_$model$($y$\_$test, xgb$\_$predictions$)$

print$($f$"$XGBoost Regressor: MAE$=\{$xgb$\_$metrics$[0]$:$\mathrm{.}4$f$\},$ MSE$=\{$xgb$\_$metrics$[1]$:$\mathrm{.}4$f$\},$ RMSE$=\{$xgb$\_$metrics$[2]$:$\mathrm{.}4$f$\},$ R$2=\{$xgb$\_$metrics$[3]$:$\mathrm{.}4$f$\}")$

#Result printed: fitting $3$ folds for each of $324$ candidates, totalling $972$ fits

Best Parameters: $\{\text{'}$colsample$\_$bytree': $0\mathrm{.}8,$ 'learning$\_$rate': $0\mathrm{.}2,$ 'max$\_$depth': $5,\text{'}$n$\_$estimators': $300,$ 'subsample': $0\mathrm{.}9\}$

XGBoost Regressor: MAE$=0\mathrm{.}3646,$ MSE$=0\mathrm{.}3808,$ RMSE$=0\mathrm{.}6171,$ R$2=0\mathrm{.}6167$