Data Preparation (18) - Prepare (Numeric and Categoric Variables)
numeric, categoric 변수에 대한 확인도 필요하다.
여기서는 numeric, categoric 모두 이름과 인덱스를 통해 확인한다.
인덱스를 사용하는 것은 해당 데이터세트내에 변수를 항상 일정한 순서로 배열하며, 모든 변수가 존재한다고 가정해야함을 의미한다.
> which(sapply(ds, is.numeric))
min_temp max_temp rainfall evaporation
3 4 5 6
sunshine wind_gust_speed wind_speed_9am wind_speed_3pm
7 9 12 13
humidity_9am humidity_3pm pressure_9am pressure_3pm
14 15 16 17
cloud_9am cloud_3pm temp_9am temp_3pm
18 19 20 21
risk_mm
23
> intersect(inputi, which(sapply(ds, is.numeric)))
[1] 3 4 5 6 7 9 12 13 14 15 17 18 19 23
> numi <- intersect(inputi, which(sapply(ds, is.numeric)))
> numi
[1] 3 4 5 6 7 9 12 13 14 15 17 18 19 23
> names(ds)[numi] // names() : 해당 오브젝트의 이름을 정하거나 변경할 때
[1] "min_temp" "max_temp" "rainfall"
[4] "evaporation" "sunshine" "wind_gust_speed"
[7] "wind_speed_9am" "wind_speed_3pm" "humidity_9am"
[10] "humidity_3pm" "pressure_3pm" "cloud_9am"
[13] "cloud_3pm" "risk_mm"
> numc <- names(ds)[numi]
> numc
[1] "min_temp" "max_temp" "rainfall"
[4] "evaporation" "sunshine" "wind_gust_speed"
[7] "wind_speed_9am" "wind_speed_3pm" "humidity_9am"
[10] "humidity_3pm" "pressure_3pm" "cloud_9am"
[13] "cloud_3pm" "risk_mm"
> which(sapply(ds, is.factor))
location wind_gust_dir wind_dir_9am wind_dir_3pm rain_today
2 8 10 11 22
rain_tomorrow
24
> intersect(inputi, which(sapply(ds, is.factor)))
[1] 8 10 11 22
> cati <- intersect(inputi, which(sapply(ds, is.factor)))
> cati
[1] 8 10 11 22
> names(ds)[cati]
[1] "wind_gust_dir" "wind_dir_9am" "wind_dir_3pm" "rain_today"
> catc <- names(ds)[cati]
> catc
[1] "wind_gust_dir" "wind_dir_9am" "wind_dir_3pm" "rain_today"
>