20.2 Decision Trees Modelling Setup

20200815

For the rattle::weatherAUS dataset we similarly define the following template variables (Graham J. Williams 2017) used for predictive modelling. See Chapter 8 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  

The 275,410 observations from the dataset have been randomly partitioned into a training dataset with 192,787 observations, a tuning dataset with 41,311 observations, and a testing dataset with 41,312 observations. The target variable (rain_tomorrow) has the classes: No (216486), Yes (58924).

References

Williams, Graham J. 2017. The Essentials of Data Science: Knowledge Discovery Using r. The r Series. CRC Press.

---

If you find this curated material useful then you can consider a donation to support it's ongoing availability and give you access to the PDF version of this book. The material has been scoped up by Generative AI without permission or any kind of recompense so do consider a donation if you can afford it. Unlike Generative AI your access to this materials is freely given. Desktop Survival Guides include Data Science, GNU/Linux, and MLHub. Books available on Amazon include Data Mining with Rattle and Essentials of Data Science. Togaware has a 30 year tradition of making popular open source software which includes sold privacy preserving productivity apps, rattle, wajig, and mlhub. Hosted by Togaware, a pioneer of free and open source software since 1984. Copyright © 1995-2022 Graham.Williams@togaware.com Creative Commons Attribution-ShareAlike 4.0