| from sklearn.base import BaseEstimator, TransformerMixin | |
| from sklearn.impute import SimpleImputer | |
| import re | |
| class PrepProcesor(BaseEstimator, TransformerMixin): | |
| def fit(self, X, y=None): | |
| self.ageImputer = SimpleImputer() | |
| self.ageImputer.fit(X[['Age']]) | |
| return self | |
| def transform(self, X, y=None): | |
| X['Age'] = self.ageImputer.transform(X[['Age']]) | |
| X['CabinClass'] = X['Cabin'].fillna('M').apply(lambda x: str(x).replace(" ", "")).apply(lambda x: re.sub(r'[^a-zA-Z]', '', x)) | |
| X['CabinNumber'] = X['Cabin'].fillna('M').apply(lambda x: str(x).replace(" ", "")).apply(lambda x: re.sub(r'[^0-9]', '', x)).replace('', 0) | |
| X['Embarked'] = X['Embarked'].fillna('M') | |
| X = X.drop(['PassengerId', 'Name', 'Ticket','Cabin'], axis=1) | |
| return X | |
| columns = ['PassengerId', 'Pclass', 'Name', 'Sex', 'Age', 'SibSp', 'Parch','Ticket', 'Fare', 'Cabin', 'Embarked'] |