Data Science Desktop Survival Guide by Graham Williams
Desktop Survival Project Home Preface Data Science Introducing R R Constructs R Tasks R Strings R Read, Write, and Create Data Template Data Exploration Data Wrangling Data Visualisation Statistics ML Template ML Scenarios ML Activities ML Applications ML Algorithms Cluster Analysis Decision Trees Computer Vision Graph Data Privacy Literate Data Science Coding with Style Resources Bibliography Index

CLICK HERE TO VISIT THE UPDATED SURVIVAL GUIDE

Decision Trees Modelling Setup

20200815

The rattle::weatherAUS dataset suggests the following the template variables (Williams, 2017) for predictive modelling. See Chapter 7 for details.

risk   <- "risk_mm"
id     <- c("date", "location")
ignore <- c(risk, id)
vars   <- setdiff(vars, ignore)
inputs <- setdiff(vars, target)

form   <- formula(target %s+% " ~ .")

ds[vars] %<>% na.roughfix()

SPLIT <- c(0.70, 0.15, 0.15)

nobs %>% sample(SPLIT[1]*nobs)                               -> tr
nobs %>% seq_len() %>% setdiff(tr) %>% sample(SPLIT[2]*nobs) -> tu
nobs %>% seq_len() %>% setdiff(tr) %>% setdiff(tu)           -> te

ds %>% slice(tr) %>% pull(target) -> actual_tr
ds %>% slice(tu) %>% pull(target) -> actual_tu
ds %>% slice(te) %>% pull(target) -> actual_te

ds %>% slice(tr) %>% pull(risk) -> risk_tr
ds %>% slice(tu) %>% pull(risk) -> risk_tu
ds %>% slice(te) %>% pull(risk) -> risk_te

## # A tibble: 2 x 2
##   rain_tomorrow  Count
## * <fct>          <int>
## 1 No            139670
## 2 Yes            37077

## Error in eval(parse(text = text, keep.source = FALSE), envir): object '.' not found

The 176,747 observations from the dataset have been randomly partitioned into a training dataset with 123,722 observations, a tuning dataset with 26,512 observations, and a testing dataset with 26,513 observations. The target variable (rain_tomorrow) has the classes: No (139670), Yes (37077).