log_reg_penguin data
Kelvin Nyongesa
2022-06-28
We will classify penguins using a simple logistics regression
steps we will follow 1. load the data set 2. Clean the data 3.Train and test 4. create a model 5. Evaluate the model ##1. load the data Lets get started by loading the libraries we will be using
library(tidyverse)## ── Attaching packages ─────────────────────────────────────── tidyverse 1.3.1 ──## ✔ ggplot2 3.3.6 ✔ purrr 0.3.4
## ✔ tibble 3.1.7 ✔ dplyr 1.0.9
## ✔ tidyr 1.2.0 ✔ stringr 1.4.0
## ✔ readr 2.1.2 ✔ forcats 0.5.1## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ dplyr::filter() masks stats::filter()
## ✖ dplyr::lag() masks stats::lag()library(palmerpenguins)
penguins## # A tibble: 344 × 8
## species island bill_length_mm bill_depth_mm flipper_length_mm body_mass_g
## <fct> <fct> <dbl> <dbl> <int> <int>
## 1 Adelie Torgersen 39.1 18.7 181 3750
## 2 Adelie Torgersen 39.5 17.4 186 3800
## 3 Adelie Torgersen 40.3 18 195 3250
## 4 Adelie Torgersen NA NA NA NA
## 5 Adelie Torgersen 36.7 19.3 193 3450
## 6 Adelie Torgersen 39.3 20.6 190 3650
## 7 Adelie Torgersen 38.9 17.8 181 3625
## 8 Adelie Torgersen 39.2 19.6 195 4675
## 9 Adelie Torgersen 34.1 18.1 193 3475
## 10 Adelie Torgersen 42 20.2 190 4250
## # … with 334 more rows, and 2 more variables: sex <fct>, year <int>###Lets ow do a simple visualize to see how our data set will look
ggplot(penguins,aes(flipper_length_mm, bill_length_mm, color = sex, size = body_mass_g)) +
geom_point(alpha = 0.5) +
facet_wrap(~species)## Warning: Removed 2 rows containing missing values (geom_point).
##2. Clealing the data Given our model will use sex for classification
we will drop the missing values. Then drop the data variables we might
not in the classification
penguins_df <- penguins %>%
filter(!is.na(sex)) %>%
select(-year, -island)
penguins_df## # A tibble: 333 × 6
## species bill_length_mm bill_depth_mm flipper_length_mm body_mass_g sex
## <fct> <dbl> <dbl> <int> <int> <fct>
## 1 Adelie 39.1 18.7 181 3750 male
## 2 Adelie 39.5 17.4 186 3800 female
## 3 Adelie 40.3 18 195 3250 female
## 4 Adelie 36.7 19.3 193 3450 female
## 5 Adelie 39.3 20.6 190 3650 male
## 6 Adelie 38.9 17.8 181 3625 female
## 7 Adelie 39.2 19.6 195 4675 male
## 8 Adelie 41.1 17.6 182 3200 female
## 9 Adelie 38.6 21.2 191 3800 male
## 10 Adelie 34.6 21.1 198 4400 male
## # … with 323 more rows##3. Training and testing Here the sex variables been our stata is splited into traning and testing
library(tidymodels)## ── Attaching packages ────────────────────────────────────── tidymodels 0.2.0 ──## ✔ broom 0.8.0 ✔ rsample 1.0.0
## ✔ dials 1.0.0 ✔ tune 0.2.0
## ✔ infer 1.0.2 ✔ workflows 0.2.6
## ✔ modeldata 0.1.1 ✔ workflowsets 0.2.1
## ✔ parsnip 1.0.0 ✔ yardstick 1.0.0
## ✔ recipes 0.2.0## ── Conflicts ───────────────────────────────────────── tidymodels_conflicts() ──
## ✖ scales::discard() masks purrr::discard()
## ✖ dplyr::filter() masks stats::filter()
## ✖ recipes::fixed() masks stringr::fixed()
## ✖ dplyr::lag() masks stats::lag()
## ✖ yardstick::spec() masks readr::spec()
## ✖ recipes::step() masks stats::step()
## • Learn how to get started at https://www.tidymodels.org/start/set.seed(123)
penguin_split <- initial_split(penguins_df, strata = sex)
penguin_train <- training(penguin_split)
penguin_test <- testing(penguin_split)we need to evaluate our model by checking on the trained data.
hence we will use bootstraps to help us evaluate now let’s create bootstrap resamples of the training data, to evaluate our models.
set.seed(123)
penguin_boot <- bootstraps(penguin_train)
penguin_boot## # Bootstrap sampling
## # A tibble: 25 × 2
## splits id
## <list> <chr>
## 1 <split [249/93]> Bootstrap01
## 2 <split [249/91]> Bootstrap02
## 3 <split [249/90]> Bootstrap03
## 4 <split [249/91]> Bootstrap04
## 5 <split [249/85]> Bootstrap05
## 6 <split [249/87]> Bootstrap06
## 7 <split [249/94]> Bootstrap07
## 8 <split [249/88]> Bootstrap08
## 9 <split [249/95]> Bootstrap09
## 10 <split [249/89]> Bootstrap10
## # … with 15 more rows##4. Building the model lets set our logistic regression model
glm_spec <- logistic_reg() %>%
set_engine("glm")
glm_spec## Logistic Regression Model Specification (classification)
##
## Computational engine: glmlet’s start putting together a tidymodels workflow(), a helper object to help manage modeling pipelines with pieces that fit together.
penguin_wf <- workflow() %>%
add_formula(sex ~ .)
penguin_wf## ══ Workflow ════════════════════════════════════════════════════════════════════
## Preprocessor: Formula
## Model: None
##
## ── Preprocessor ────────────────────────────────────────────────────────────────
## sex ~ .Lets now add a model, and the fit to each of the resamples. First, we can fit the logistic regression model
glm_rs <- penguin_wf %>%
add_model(glm_spec) %>%
fit_resamples(
resamples = penguin_boot,
control = control_resamples(save_pred = TRUE)
)## ! Bootstrap05: preprocessor 1/1, model 1/1: glm.fit: fitted probabilities numerically 0...## ! Bootstrap08: preprocessor 1/1, model 1/1: glm.fit: fitted probabilities numerically 0...## ! Bootstrap23: preprocessor 1/1, model 1/1: glm.fit: fitted probabilities numerically 0...glm_rs## # Resampling results
## # Bootstrap sampling
## # A tibble: 25 × 5
## splits id .metrics .notes .predictions
## <list> <chr> <list> <list> <list>
## 1 <split [249/93]> Bootstrap01 <tibble [2 × 4]> <tibble [0 × 3]> <tibble>
## 2 <split [249/91]> Bootstrap02 <tibble [2 × 4]> <tibble [0 × 3]> <tibble>
## 3 <split [249/90]> Bootstrap03 <tibble [2 × 4]> <tibble [0 × 3]> <tibble>
## 4 <split [249/91]> Bootstrap04 <tibble [2 × 4]> <tibble [0 × 3]> <tibble>
## 5 <split [249/85]> Bootstrap05 <tibble [2 × 4]> <tibble [1 × 3]> <tibble>
## 6 <split [249/87]> Bootstrap06 <tibble [2 × 4]> <tibble [0 × 3]> <tibble>
## 7 <split [249/94]> Bootstrap07 <tibble [2 × 4]> <tibble [0 × 3]> <tibble>
## 8 <split [249/88]> Bootstrap08 <tibble [2 × 4]> <tibble [1 × 3]> <tibble>
## 9 <split [249/95]> Bootstrap09 <tibble [2 × 4]> <tibble [0 × 3]> <tibble>
## 10 <split [249/89]> Bootstrap10 <tibble [2 × 4]> <tibble [0 × 3]> <tibble>
## # … with 15 more rows
##
## There were issues with some computations:
##
## - Warning(s) x3: glm.fit: fitted probabilities numerically 0 or 1 occurred
##
## Use `collect_notes(object)` for more information.##5. Evaluate the model
collect_metrics(glm_rs)## # A tibble: 2 × 6
## .metric .estimator mean n std_err .config
## <chr> <chr> <dbl> <int> <dbl> <chr>
## 1 accuracy binary 0.918 25 0.00639 Preprocessor1_Model1
## 2 roc_auc binary 0.979 25 0.00254 Preprocessor1_Model1###Now you have created your first working lOGISTIC REGRESSION MODEL