house_prices.py 文件源码

def drop_variable_before_preparation(self, df):
        # Acceptable limit of NaN in features
        limit_of_nans = 0.3*df.shape[0]
        # limit_of_nans = 800
        for feature in self.features_with_missing_values_in_dataframe(df).index:
            if df[feature].isnull().sum() > limit_of_nans:
                df = df.drop([feature], axis=1)

        # df = df.drop(['Alley'], axis=1)
        # df = df.drop(['MasVnrType'], axis=1)
        # df = df.drop(["Utilities", "LotFrontage", "Alley", "MasVnrType", "MasVnrArea", "BsmtQual",
        #               "BsmtCond", "BsmtExposure", "BsmtFinType1", "BsmtFinType2",
        #               "Electrical", "FireplaceQu", "GarageType", "GarageYrBlt",
        #               "GarageFinish", "GarageQual", "GarageCond", "PoolQC",
        #               "Fence", "MiscFeature"], axis=1)
        return df