-
Notifications
You must be signed in to change notification settings - Fork 0
/
transform.py
171 lines (130 loc) · 7.07 KB
/
transform.py
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
import numpy as np
from scipy import stats
# Get the original dataframe and transform the features
# according to the rules defined
def transform_dataset(house_prices):
y = np.log(house_prices['SalePrice'])
x = house_prices.drop(columns=['SalePrice'])
# Basement variables
x['BsmtQualBinary'] = (x['BsmtQual'] == 'Ex') | (x['BsmtQual'] == 'Gd')
x['BsmtExposureBinary'] = (x['BsmtExposure'] == 'Gd') | (x['BsmtExposure'] == 'Av') | (x['BsmtExposure'] == 'Mn')
x.drop(columns=['BsmtQual', 'BsmtExposure'], inplace=True)
# Fireplace variables
x['FireplaceQualBinary'] = (x['FireplaceQu'] == 'Ex') | (x['FireplaceQu'] == 'Gd') | (x['FireplaceQu'] == 'TA')
x.drop(columns=['FireplaceQu'], inplace=True)
# Garage variables
x['GarageTypeBinary'] = (x['GarageType'] == 'Attchd') | (x['GarageType'] == 'BuiltIn')
x['GarageFinishBinary'] = (x['GarageFinish'] == 'Fin') | (x['GarageFinish'] == 'RFn')
x.drop(columns=['GarageType', 'GarageFinish'], inplace=True)
# Electrical variables
x['ElectricalSBrkr'] = x['Electrical'] == 'SBrkr'
x.drop(columns=['Electrical'], inplace=True)
# Masonry veneer variables
x['MasVnrTypeBinary'] = (x['MasVnrType'] == 'BrkFace') | (x['MasVnrType'] == 'Stone')
x.drop(columns=['MasVnrType'], inplace=True)
# Paved Drive variable
x['PavedDriveBinary'] = x['PavedDrive'] == 'Y'
x.drop(columns=['PavedDrive'], inplace=True)
# Sale condition variables
x['SaleConditionBinary1'] = x['SaleCondition'] == 'Partial'
x['SaleConditionBinary2'] = (x['SaleCondition'] == 'Abnorml') | (x['SaleCondition'] == 'AdjLand')
x.drop(columns=['SaleCondition'], inplace=True)
# Sale type variables
x['SaleTypeBinary1'] = x['SaleType'] == 'New'
x['SaleTypeBinary2'] = (x['SaleType'] == 'Oth') | (x['SaleType'] == 'ConLD') | (x['SaleType'] == 'COD')
x.drop(columns=['SaleType'], inplace=True)
# Kitchen quality variable
x['KitchenQualBinary1'] = (x['KitchenQual'] == 'Ex') | (x['KitchenQual'] == 'Gd')
x.drop(columns=['KitchenQual'], inplace=True)
# Foundation variable
x['FoundationBinary'] = x['Foundation'] == 'PConc'
x.drop(columns=['Foundation'], inplace=True)
# Exterior condition variable
x['ExterCondBinary'] = (x['ExterCond'] == 'TA') | (x['ExterCond'] == 'Gd') | (x['ExterCond'] == 'Ex')
x.drop(columns=['ExterCond'], inplace=True)
# Exterior quality variable
x['ExterQualBinary'] = (x['ExterQual'] == 'Ex') | (x['ExterQual'] == 'Gd')
x.drop(columns=['ExterQual'], inplace=True)
# General zoning classification of sale variables
x['MSZoningBinary1'] = x['MSZoning'] == 'FV'
x['MSZoningBinary2'] = (x['MSZoning'] == 'C (all)') | (x['MSZoning'] == 'RH') | (x['MSZoning'] == 'RM')
x.drop(columns=['MSZoning'], inplace=True)
# Roof style variable
x['RoofStyleBinary'] = x['RoofStyle'] == 'Hip'
x.drop(columns=['RoofStyle'], inplace=True)
# Overall quality variable
x['OverallQualBinary'] = x['OverallQual'].apply(lambda x: int(x) > 5)
x.drop(columns=['OverallQual'], inplace=True)
# Neighborhood variables
x['NeighborhoodPoor'] = ((x['Neighborhood'] == 'BrDale') | (x['Neighborhood'] == 'BrkSide') | (x['Neighborhood'] == 'Edwards') |
(x['Neighborhood'] == 'IDOTRR') | (x['Neighborhood'] == 'MeadowV') | (x['Neighborhood'] == 'Names') |
(x['Neighborhood'] == 'OldTown'))
x['NeighborhoodRich'] = ((x['Neighborhood'] == 'Blmngtn') | (x['Neighborhood'] == 'CollgCr') | (x['Neighborhood'] == 'Gilbirt') |
(x['Neighborhood'] == 'NoRidge') | (x['Neighborhood'] == 'NWAmes') | (x['Neighborhood'] == 'NridgHt') |
(x['Neighborhood'] == 'Somerst') | (x['Neighborhood'] == 'StoneBr') | (x['Neighborhood'] == 'Timber') |
(x['Neighborhood'] == 'Veenker'))
x.drop(columns=['Neighborhood'], inplace=True)
# Type of dwelling involved in the sale variables
x['MSSubClassBinary1'] = ((x['MSSubClass'] == 30) | (x['MSSubClass'] == 45) | (x['MSSubClass'] == 50) |
(x['MSSubClass'] == 90) | (x['MSSubClass'] == 180) | (x['MSSubClass'] == 190))
x['MSSubClassBinary2'] = (x['MSSubClass'] == 60) | (x['MSSubClass'] == 120)
x.drop(columns=['MSSubClass'], inplace=True)
# Overall condition variable
x['OverallCondBinary'] = x['OverallCond'] > 4
x.drop(columns=['OverallCond'], inplace=True)
# House style variables
x['HouseStyleBinary1'] = (x['HouseStyle'] == '1.5Fin') | (x['HouseStyle'] == '1.5Unf')
x['HouseStyleBinary2'] = (x['HouseStyle'] == '2Story')
x.drop(columns=['HouseStyle'], inplace=True)
# Exterior covering on house variable
x['Exterior1stBinary1'] = ((x['Exterior1st'] == 'AsbShng') | (x['Exterior1st'] == 'AsphShn') | (x['Exterior1st'] == 'BrkComm') |
(x['Exterior1st'] == 'CBlock') | (x['Exterior1st'] == 'MetalSd') | (x['Exterior1st'] == 'Wd Sdng') |
(x['Exterior1st'] == 'WdShing'))
x['Exterior1stBinary2'] = (x['Exterior1st'] == 'ImStucc') | (x['Exterior1st'] == 'Stone') | (x['Exterior1st'] == 'VinylSd')
x.drop(columns=['Exterior1st'], inplace=True)
# Lot variables
x['LotShapeBinary'] = x['LotShape'] == 'Reg'
x['LotConfigBinary'] = x['LotConfig'] == 'CulDSac'
x.drop(columns=['LotShape', 'LotConfig'], inplace=True)
# Basement variable
x['BsmtFinType1Binary'] = x['BsmtFinType1'] == 'GLQ'
x.drop(columns=['BsmtFinType1'], inplace=True)
# Heating/Central Air variables
x['HeatingQCBinary'] = x['HeatingQC'] == 'Ex'
x['CentralAirBinary'] = x['CentralAir'] == 'N'
x.drop(columns=['HeatingQC', 'CentralAir'], inplace=True)
x.drop(columns=['BsmtFullBath', 'BsmtHalfBath'], inplace=True)
# Drop some variables
x.drop(columns=['Street', 'Condition2', 'Condition1', 'RoofMatl', 'Heating', 'LandSlope', 'LandContour', 'BldgType',
'Functional', 'Exterior2nd', 'MiscVal', 'Utilities'], inplace=True)
# Lot size variable
x['LotArea'] = np.log(x['LotArea'])
# Ground living area variable
x['Has2ndFlr'] = x['2ndFlrSF'] > 0
x.drop(columns=['1stFlrSF', '2ndFlrSF', 'LowQualFinSF'], inplace=True)
# Basement variables
x.drop(columns=['BsmtFinSF1', 'BsmtFinSF2', 'BsmtUnfSF'], inplace=True)
# Lot frontage variable
trans, lam = stats.boxcox(x['LotFrontage'])
x['LotFrontage'] = trans
# Garage area variable
trans, lam = stats.boxcox(x['GarageArea'] + 1)
x['GarageArea'] = trans
# Drop some continuous variables
x.drop(columns=['EnclosedPorch', '3SsnPorch', 'ScreenPorch', 'PoolArea'], inplace=True)
# Continuous countable variables
x['GarageCarsBinary'] = x['GarageCars'].apply(lambda x: int(x) > 1)
x['FireplacesBinary'] = x['Fireplaces'].apply(lambda x: int(x) > 0)
x['TotRmsAbvGrdBinary1'] = x['TotRmsAbvGrd'].apply(lambda x: int(x) > 5 and int(x) <= 7)
x['TotRmsAbvGrdBinary2'] = x['TotRmsAbvGrd'].apply(lambda x: int(x) > 7)
x['KitchenAbvGrBinary'] = x['KitchenAbvGr'].apply(lambda x: int(x) == 1)
# x['FullBathBinary'] = x['FullBath'].apply(lambda x: int(x) > 1)
# x['HalfBathBinary'] = x['HalfBath'].apply(lambda x: int(x) == 1)
x['BedroomAbvGrBinary'] = x['BedroomAbvGr'].apply(lambda x: int(x) == 4)
x.drop(columns=['GarageCars', 'Fireplaces', 'TotRmsAbvGrd', 'KitchenAbvGr',
'FullBath', 'HalfBath', 'BedroomAbvGr'], inplace=True)
# Continuous year / age variables
x['NewHouse'] = x['YearBuilt'] > 1990
x['OldHouse'] = x['YearBuilt'] < 1980
x.drop(columns=['YearRemodAdd', 'MoSold', 'YrSold'], inplace=True)
return x, y