presRandomForests/main.R at master · suvasama/presRandomForests · GitHub

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
# LOAD DATA

library(data.table)
library(here)

path = here("Data",
            "cal_housing.data")

col_names = c("longitude",
              "latitude",
              "housingMedianAge",
              "totalRooms",
              "totalBedrooms",
              "population",
              "households",
              "medianIncome",
              "medianHouseValue")

data <- fread(path,
              col.names = col_names)

rm(path)

##-------------------------------------------------------------------------------------

# FEATURE ENGINEERING

cols <- c("totalRooms",
          "totalBedrooms")

setnames(
  data[,
       (cols) := lapply(.SD, "/", data$population),
       .SDcols = cols],
         cols,
  c("avg_rooms",
    "avg_beds")
  )

setnames(
  data[, population := population / households],
  "population",
  "avg_population"
  )

cols <- c("medianHouseValue",
          "housingMedianAge",
          "households",
          "avg_population",
          "avg_rooms",
          "avg_beds")

new_cols <- c("log_house_value",
              "log_age",
              "log_households",
              "log_avg_population",
              "log_avg_rooms",
              "log_avg_beds")

setnames(
  data[,
       (cols) := lapply(.SD, log),
       .SDcols = cols],
  cols,
  new_cols
  )

data[,
     `:=` (
       med_income2 = medianIncome^2,
       med_income3 = medianIncome^3
       )
     ]

y <- data$log_house_value
X <- data
X$log_house_value <- NULL

# indices for train-test-split
set.seed(101)
train <- sample(1:nrow(data),
                floor(0.75*nrow(data))
                )

prices_test = data[-train,
                   .(log_house_value)]

rm(cols)
rm(new_cols)