X$\_$train, X$\_$test, y$\_$train, y$\_$test $=$ train$\_$test$\_$split $,$ test$\_$size $=0\mathrm{.}2,$ random$\_$state $=42$

# Define the column transformer for preprocessing

numeric$\_$features $=[\text{'}$YrSold$\text{'},$ 'SaleType'$]$

categorical$\_$features $=[\text{'}$GarageCars$\text{'},\text{'}$ PoolArea'$]$

preprocessor $=$ ColumnTransformer$($

transformers

$(\text{'}$num$\text{'},$ MinMaxScaler$(),$ numeric$\_$features$),$

$(\text{'}$cat$\text{'},$ OneHotEncoder$(),$ categorical$\_$features$)$

# Create Random Forest pipeline

rf$\_$pipeline $=$ Pipeline$($steps$=[(\text{'}$preprocessor$\text{'},$ preprocessor$),$

$(\text{'}$regressor$\text{'},$ RandomForestRegressor$())])$

# Create K$-$Nearest Neighbors pipeline

knn$\_$pipeline $=$ Pipeline$($steps$=[(\text{'}$preprocessor$\text{'},$ preprocessor$),$

$(\text{'}$regressor$\text{'},$ KNeighborsRegressor$())])$

# Define hyperparameters grid for GridSearchcv

rf$\_$param$\_$grid

'regressor$\_$n$\_$estimators $\text{'}$: $[50,100],$

'regressor$\_$max$\_$depth': $[2,3,4],$

$\}$

'regressor$\_$criterion': $[\text{'}$mse$\text{'},$ 'mae'$]$

knn$\_$param$\_$grid

'regressor$\_$n$\_$neighbors': $[2,5,10,20,50,100],$

$\}$

'regressor$\_$weights': $[\text{'}$uniform$\text{'},$ 'distance'$]$

# Perform GridSearchCV with $5-$fold cross validation for Random Forest

rf$\_$grid$\_$search $=$ GridSearchCV$($rfR

Calculate the following evaluation metrics of the $2$ models' performance on the training data set:

Root mean squared error MSE

Mean absolute error MAE

Mean absolute percentage error MAPE

Write a paragraph describing the results and which model and set of hyper parameters worked the best and based on which accuracy metric$($s$)?$ If you were to explore more hyper parameters for each model, how would you expand or limit the current hyperparameter grid.

as

$[19]$ df $=$ pd$.$read$\_$csv$(\text{'}$train$.$csv$\text{'})$

$[20]$ df$.$columns

Index $(\text{'}$Id$\text{'},$ 'MSSubClass', 'MSZoning', 'LotFrontage', 'LotArea', 'Street',

'Alley', 'LotShape', 'LandContour', 'Utilities', 'LotConfig',

'LandSlope', 'Neighborhood', 'Condition$1\text{'},$ 'Condition$2\text{'},$ 'BldgType',

'HouseStyle', 'Overallqual', 'Overalicond', 'YearBuilt', 'YearRemodAdd',

'RoofStyle', 'RoofMatl', 'Exterior$1$st$\text{'},$ 'Exterior$2$nd$\text{'},$ 'MasVnrType',

'MasVnrArea', 'ExterQual': 'ExterCond', 'Foundation', 'BsmtQual',

'BsmtCond', 'BsmtExposure', 'BsmtFinType$1\text{'},$ 'BsmtFinSF$1\text{'},\_$pipeline, rf$\_$paran$\_$grid, cv$=5)$