05_tidymodels.Rmd

---
title: "tidymodels"
author: "Axel R"
date: "2024-01-21"
output: html_document
---

```{r message=FALSE, warning=FALSE}
DataExplorer::create_report(taxi)
library(parsnip)
```

# Data Budget

## The initial SPlit

```{r}
set.seed(123)
taxi_split <- initial_split(taxi)
taxi_split
```

## Accessing the data

```{r}
taxi_train <- training(taxi_split)
taxi_test <- testing(taxi_split)
```

## Data splitting and spending

```{r}
set.seed(123)
taxi_split <- initial_split(taxi, prop = 0.8)
taxi_train <- training(taxi_split)
taxi_test <- testing(taxi_split)

nrow(taxi_train)
#> [1] 8000
nrow(taxi_test)
#> [1] 2000
```

## Validation set

```{r}
set.seed(123)
initial_validation_split(taxi, prop = c(0.6, 0.2))
```

## Stratification

```{r}
set.seed(123)
taxi_split <- initial_split(taxi, prop = 0.8, strata = tip)
taxi_split
```

```{r}
# Model
logistic_reg()
# Engine
logistic_reg() %>%
  set_engine("glm")
# 
decision_tree() %>% 
  set_mode("classification")
```

## Model Workflow

```{r}
tree_spec <-
  decision_tree(cost_complexity = 0.002) %>% 
  set_mode("classification")

workflow() %>%
  add_formula(tip ~ .) %>%
  add_model(tree_spec) %>%
  fit(data = taxi_train)
```