VV_metrische_Variablen.Rmd

---
title: '7.2'
author: "Emilia Braun"
date: "2023-04-26"
output: html_document
---
yeo johnson

macht schiefe Verteilungen wieder gerade
```{r}
data(biomass, package = "modeldata")

biomass_tr <- biomass[biomass$dataset == "Training", ]
biomass_te <- biomass[biomass$dataset == "Testing", ]

rec <- recipe(
  HHV ~ carbon + hydrogen + oxygen + nitrogen + sulfur,
  data = biomass_tr
)

yj_transform <- step_YeoJohnson(rec, all_numeric())

yj_estimates <- prep(yj_transform, training = biomass_tr)

yj_te <- bake(yj_estimates, biomass_te)
```

```{r}
plot(density(biomass_te$sulfur), main = "before")
```


```{r}
plot(density(yj_te$sulfur), main = "after")
```

step_nzv

schmeißt Spalten mit near zero variance aus dem Datensatz.
in diesem Fall gibt es eine Spalte "sparse", die nur aus dem Wert 0 besteht, mit der Ausnahme der ersten Zeile, die den Wert 1 hat
```{r}
data(biomass, package = "modeldata")

biomass$sparse <- c(1, rep(0, nrow(biomass) - 1))

biomass_tr <- biomass[biomass$dataset == "Training", ]
biomass_te <- biomass[biomass$dataset == "Testing", ]

rec <- recipe(HHV ~ carbon + hydrogen + oxygen +
  nitrogen + sulfur + sparse,
data = biomass_tr
)

nzv_filter <- rec %>%
  step_nzv(all_predictors())

filter_obj <- prep(nzv_filter, training = biomass_tr)

filtered_te <- bake(filter_obj, biomass_te)
any(names(filtered_te) == "sparse")

# tidy zeigt an, welche Spalte entfernt wird
tidy(nzv_filter, number = 1)
tidy(filter_obj, number = 1)

```
step_zv

schmeißt Spalten raus, die zero variance haben, also eine Spalte in der nur ein Wert ist
in diesem Fall die Spalte "one_value", die in jeder Zelle den Wert 1 stehen hat


```{r}
data(biomass, package = "modeldata")

biomass$one_value <- 1

biomass_tr <- biomass[biomass$dataset == "Training", ]
biomass_te <- biomass[biomass$dataset == "Testing", ]

rec <- recipe(HHV ~ carbon + hydrogen + oxygen +
  nitrogen + sulfur + one_value,
data = biomass_tr
)

zv_filter <- rec %>%
  step_zv(all_predictors())

filter_obj <- prep(zv_filter, training = biomass_tr)

filtered_te <- bake(filter_obj, biomass_te)
any(names(filtered_te) == "one_value")

# tidy zeigt an, welche Spalte entfernt wird
tidy(zv_filter, number = 1)
tidy(filter_obj, number = 1)

```