I have a dataset called climbers_data that looks like this.
climbers_data %>% glimpse() ## Rows: 76,519
## Columns: 21
## $ expedition_id <chr> "AMAD78301", "AMAD78301", "AMAD78301", "AMAD78301…
## $ member_id <chr> "AMAD78301-01", "AMAD78301-02", "AMAD78301-03", "…
## $ peak_id <chr> "AMAD", "AMAD", "AMAD", "AMAD", "AMAD", "AMAD", "…
## $ peak_name <chr> "Ama Dablam", "Ama Dablam", "Ama Dablam", "Ama Da…
## $ year <dbl> 1978, 1978, 1978, 1978, 1978, 1978, 1978, 1978, 1…
## $ season <chr> "Autumn", "Autumn", "Autumn", "Autumn", "Autumn",…
## $ sex <chr> "M", "M", "M", "M", "M", "M", "M", "M", "M", "M",…
## $ age <dbl> 40, 41, 27, 40, 34, 25, 41, 29, 35, 37, 23, 44, 2…
## $ citizenship <chr> "France", "France", "France", "France", "France",…
## $ expedition_role <chr> "Leader", "Deputy Leader", "Climber", "Exp Doctor…
## $ hired <lgl> FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, …
## $ highpoint_metres <dbl> NA, 6000, NA, 6000, NA, 6000, 6000, 6000, NA, 681…
## $ success <lgl> FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, …
## $ solo <lgl> FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, …
## $ oxygen_used <lgl> FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, …
## $ died <lgl> FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, …
## $ death_cause <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
## $ death_height_metres <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
## $ injured <lgl> FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, FALSE, …
## $ injury_type <chr> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…
## $ injury_height_metres <dbl> NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, NA, N…Rows: 52,383 Columns: 17 $ expedition_id
The goal is to help predict died for members.
Please write R code to create a classification model that predicts the probability of died.
# Load libraries
library(tidyverse)
library(tidymodels)## ── Attaching packages ────────────────────────────────────── tidymodels 1.2.0 ──## ✔ broom 1.0.5 ✔ rsample 1.2.1
## ✔ dials 1.2.1 ✔ tune 1.2.1
## ✔ infer 1.0.6 ✔ workflows 1.1.4
## ✔ modeldata 1.3.0 ✔ workflowsets 1.1.0
## ✔ parsnip 1.2.1 ✔ yardstick 1.3.1
## ✔ recipes 1.0.10## ── 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()
## • Use tidymodels_prefer() to resolve common conflicts.library(h2o)##
## ----------------------------------------------------------------------
##
## Your next step is to start H2O:
## > h2o.init()
##
## For H2O package documentation, ask for help:
## > ??h2o
##
## After starting H2O, you can use the Web UI at http://localhost:54321
## For more information visit https://docs.h2o.ai
##
## ----------------------------------------------------------------------##
## Attaching package: 'h2o'## The following objects are masked from 'package:lubridate':
##
## day, hour, month, week, year## The following objects are masked from 'package:stats':
##
## cor, sd, var## The following objects are masked from 'package:base':
##
## &&, %*%, %in%, ||, apply, as.factor, as.numeric, colnames,
## colnames<-, ifelse, is.character, is.factor, is.numeric, log,
## log10, log1p, log2, round, signif, trunc# Load data
climbers_data_clean <- climbers_data
# Preprocess data
climbers_data_processed <- climbers_data %>%
select(-expedition_id, -member_id, -peak_id, -peak_name) %>% # Remove unnecessary columns
mutate(success = as.factor(success), # Convert success to factor
solo = as.factor(solo), # Convert solo to factor
oxygen_used = as.factor(oxygen_used), # Convert oxygen_used to factor
died = as.factor(died), # Convert died to factor
injured = as.factor(injured)) # Convert injured to factor
# Set seed for reproducibility
set.seed(123)
# Split data into training (80%) and testing (20%) sets
data_split <- initial_split(climbers_data_processed, prop = 0.8, strata = died)
train_data <- training(data_split)
test_data <- testing(data_split)
# Initialize h2o
h2o.init()## Connection successful!
##
## R is connected to the H2O cluster:
## H2O cluster uptime: 21 hours 46 minutes
## H2O cluster timezone: America/New_York
## H2O data parsing timezone: UTC
## H2O cluster version: 3.44.0.3
## H2O cluster version age: 4 months and 11 days
## H2O cluster name: H2O_started_from_R_Vanessa_zyd177
## H2O cluster total nodes: 1
## H2O cluster total memory: 1.30 GB
## H2O cluster total cores: 8
## H2O cluster allowed cores: 8
## H2O cluster healthy: TRUE
## H2O Connection ip: localhost
## H2O Connection port: 54321
## H2O Connection proxy: NA
## H2O Internal Security: FALSE
## R Version: R version 4.3.1 (2023-06-16)## Warning in h2o.clusterInfo():
## Your H2O cluster version is (4 months and 11 days) old. There may be a newer version available.
## Please download and install the latest version from: https://h2o-release.s3.amazonaws.com/h2o/latest_stable.html# Convert data to h2o format
train_h2o <- as.h2o(train_data)##
|
| | 0%
|
|======================================================================| 100%test_h2o <- as.h2o(test_data)##
|
| | 0%
|
|======================================================================| 100%# Define predictors and response variable
predictors <- setdiff(colnames(train_h2o), c("died"))
response <- "died"
# Train AutoML model
aml <- h2o.automl(x = predictors,
y = response,
training_frame = train_h2o,
leaderboard_frame = test_h2o,
max_runtime_secs = 30, # Set maximum runtime
seed = 123)##
|
| | 0%
|
|== | 3%
## 11:23:51.705: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
## 11:23:53.246: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|======= | 10%
## 11:23:54.651: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|============ | 17%
## 11:23:56.659: _train param, Dropping unused columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|================= | 24%
## 11:23:59.75: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|====================== | 31%
## 11:24:01.680: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|=========================== | 38%
## 11:24:02.968: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|=============================== | 45%
## 11:24:05.363: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|==================================== | 52%
## 11:24:06.820: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|========================================= | 58%
## 11:24:08.380: _train param, Dropping unused columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|============================================== | 65%
## 11:24:10.832: _train param, Dropping unused columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|================================================== | 72%
## 11:24:13.455: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|======================================================= | 79%
## 11:24:14.966: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
## 11:24:16.97: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|============================================================ | 86%
## 11:24:17.939: _train param, Dropping bad and constant columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|================================================================= | 93%
## 11:24:19.422: _train param, Dropping unused columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]
|
|======================================================================| 100%
## 11:24:21.88: _train param, Dropping unused columns: [expedition_role, season, death_cause, sex, citizenship, injury_type]# View AutoML leaderboard
print(aml@leaderboard)## model_id auc logloss
## 1 DRF_1_AutoML_6_20240502_112351 0.9977190 0.003732542
## 2 GBM_3_AutoML_6_20240502_112351 0.9974373 0.001686459
## 3 GBM_1_AutoML_6_20240502_112351 0.9973831 0.001986849
## 4 GBM_2_AutoML_6_20240502_112351 0.9972340 0.001831792
## 5 StackedEnsemble_BestOfFamily_1_AutoML_6_20240502_112351 0.9972195 0.002022247
## 6 StackedEnsemble_AllModels_2_AutoML_6_20240502_112351 0.9971773 0.001735483
## aucpr mean_per_class_error rmse mse
## 1 0.9905834 0.007352941 0.02209069 0.0004879984
## 2 0.9911733 0.007352941 0.01427139 0.0002036727
## 3 0.9905591 0.007352941 0.01616443 0.0002612889
## 4 0.9912935 0.007352941 0.01557386 0.0002425453
## 5 0.9906157 0.007352941 0.01686741 0.0002845094
## 6 0.9910759 0.007352941 0.01506485 0.0002269496
##
## [17 rows x 7 columns]# Get best model from AutoML
best_model <- aml@leader
# Make predictions on test data
predictions <- h2o.predict(best_model, test_h2o)##
|
| | 0%
|
|======================================================================| 100%# View predictions
head(predictions)## predict FALSE TRUE
## 1 FALSE 0.9996317 0.0003683484
## 2 FALSE 0.9994714 0.0005285704
## 3 FALSE 0.9989233 0.0010767376
## 4 FALSE 0.9989572 0.0010427701
## 5 FALSE 0.9997627 0.0002373469
## 6 FALSE 0.9996466 0.0003533924