Goal is to automate building and tuning a classification model to predict employee attrition, using the h2o::h2o.automl.
Import the cleaned data from Module 7.
library(h2o)## Warning: package 'h2o' was built under R version 4.3.3##
## ----------------------------------------------------------------------
##
## 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: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, trunclibrary(tidyverse)## Warning: package 'ggplot2' was built under R version 4.3.3## ── Attaching core tidyverse packages ──────────────────────── tidyverse 2.0.0 ──
## ✔ dplyr 1.1.4 ✔ readr 2.1.5
## ✔ forcats 1.0.0 ✔ stringr 1.5.1
## ✔ ggplot2 3.5.1 ✔ tibble 3.2.1
## ✔ lubridate 1.9.3 ✔ tidyr 1.3.1
## ✔ purrr 1.0.2## ── Conflicts ────────────────────────────────────────── tidyverse_conflicts() ──
## ✖ lubridate::day() masks h2o::day()
## ✖ dplyr::filter() masks stats::filter()
## ✖ lubridate::hour() masks h2o::hour()
## ✖ dplyr::lag() masks stats::lag()
## ✖ lubridate::month() masks h2o::month()
## ✖ lubridate::week() masks h2o::week()
## ✖ lubridate::year() masks h2o::year()
## ℹ Use the conflicted package (<http://conflicted.r-lib.org/>) to force all conflicts to become errorslibrary(tidymodels)## Warning: package 'tidymodels' was built under R version 4.3.3## ── Attaching packages ────────────────────────────────────── tidymodels 1.2.0 ──
## ✔ broom 1.0.5 ✔ rsample 1.2.1
## ✔ dials 1.3.0 ✔ tune 1.2.1
## ✔ infer 1.0.7 ✔ workflows 1.1.4
## ✔ modeldata 1.4.0 ✔ workflowsets 1.1.0
## ✔ parsnip 1.2.1 ✔ yardstick 1.3.1
## ✔ recipes 1.1.0## Warning: package 'dials' was built under R version 4.3.3## Warning: package 'infer' was built under R version 4.3.3## Warning: package 'modeldata' was built under R version 4.3.3## Warning: package 'parsnip' was built under R version 4.3.3## Warning: package 'recipes' was built under R version 4.3.3## Warning: package 'rsample' was built under R version 4.3.3## Warning: package 'tune' was built under R version 4.3.3## Warning: package 'workflows' was built under R version 4.3.3## Warning: package 'workflowsets' was built under R version 4.3.3## Warning: package 'yardstick' was built under R version 4.3.3## ── 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/library(tidyquant)## Loading required package: PerformanceAnalytics
## Loading required package: xts
## Loading required package: zoo
##
## Attaching package: 'zoo'
##
## The following objects are masked from 'package:base':
##
## as.Date, as.Date.numeric
##
##
## ######################### Warning from 'xts' package ##########################
## # #
## # The dplyr lag() function breaks how base R's lag() function is supposed to #
## # work, which breaks lag(my_xts). Calls to lag(my_xts) that you type or #
## # source() into this session won't work correctly. #
## # #
## # Use stats::lag() to make sure you're not using dplyr::lag(), or you can add #
## # conflictRules('dplyr', exclude = 'lag') to your .Rprofile to stop #
## # dplyr from breaking base R's lag() function. #
## # #
## # Code in packages is not affected. It's protected by R's namespace mechanism #
## # Set `options(xts.warn_dplyr_breaks_lag = FALSE)` to suppress this warning. #
## # #
## ###############################################################################
##
## Attaching package: 'xts'
##
## The following objects are masked from 'package:dplyr':
##
## first, last
##
##
## Attaching package: 'PerformanceAnalytics'
##
## The following object is masked from 'package:graphics':
##
## legend
##
## Loading required package: quantmod
## Loading required package: TTR
##
## Attaching package: 'TTR'
##
## The following object is masked from 'package:dials':
##
## momentum
##
## Registered S3 method overwritten by 'quantmod':
## method from
## as.zoo.data.frame zoodata <- read_csv("../00_data/data_wrangled/data_clean.csv") %>%
# h2o requires all variables to be either numeric or factors
mutate(across(where(is.character), factor))## Rows: 1470 Columns: 34
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (9): Attrition, BusinessTravel, Department, EducationField, Gender, Job...
## dbl (24): Age, DailyRate, DistanceFromHome, Education, EmployeeNumber, Envir...
## lgl (1): Attrition == "YES"
##
## ℹ Use `spec()` to retrieve the full column specification for this data.
## ℹ Specify the column types or set `show_col_types = FALSE` to quiet this message.set.seed(1234)
data_split <- initial_split(data, strata = "Attrition")
train_tbl <- training(data_split)
test_tbl <- testing(data_split)recipe_obj <- recipe(Attrition ~ ., data = train_tbl) %>%
# Remove zero variance variables
step_zv(all_predictors()) # Initialize h2o
h2o.init()##
## H2O is not running yet, starting it now...
##
## Note: In case of errors look at the following log files:
## C:\Users\nilss\AppData\Local\Temp\RtmpgNRzWY\file701c333c7f20/h2o_nilss_started_from_r.out
## C:\Users\nilss\AppData\Local\Temp\RtmpgNRzWY\file701c507f22d/h2o_nilss_started_from_r.err
##
##
## Starting H2O JVM and connecting: Connection successful!
##
## R is connected to the H2O cluster:
## H2O cluster uptime: 4 seconds 467 milliseconds
## H2O cluster timezone: America/New_York
## H2O data parsing timezone: UTC
## H2O cluster version: 3.44.0.3
## H2O cluster version age: 11 months and 15 days
## H2O cluster name: H2O_started_from_R_nilss_fhp551
## H2O cluster total nodes: 1
## H2O cluster total memory: 3.92 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 4.3.2 (2023-10-31 ucrt)## Warning in h2o.clusterInfo():
## Your H2O cluster version is (11 months and 15 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.htmlsplit.h2o <- h2o.splitFrame(as.h2o(train_tbl), ratio = c(0.85), seed = 2345)##
|
| | 0%
|
|======================================================================| 100%train_h2o <- split.h2o[[1]]
valid_h2o <- split.h2o[[2]]
test_h2o <- as.h2o(test_tbl)##
|
| | 0%
|
|======================================================================| 100%y <- "Attrition"
x <- setdiff(names(train_tbl), y)
models_h2o <-h2o.automl(
x = x,
y = y,
training_frame = train_h2o,
validation_frame = valid_h2o,
leaderboard_frame = test_h2o,
# max_runtime_secs = 30,
max_models = 10,
exclude_algos = "DeepLearning",
nfolds = 5,
seed = 3456
)##
|
| | 0%
## 15:52:58.739: User specified a validation frame with cross-validation still enabled. Please note that the models will still be validated using cross-validation only, the validation frame will be used to provide purely informative validation metrics on the trained models.
## 15:52:58.750: AutoML: XGBoost is not available; skipping it.
|
|= | 2%
## 15:53:03.587: GLM_1_AutoML_1_20241206_155258 [GLM def_1] failed: java.lang.ArrayIndexOutOfBoundsException: Index 55 out of bounds for length 55
|
|===== | 8%
|
|========= | 12%
|
|=============== | 21%
|
|==================== | 29%
|
|======================= | 33%
|
|=============================================== | 67%
|
|================================================== | 71%
|
|======================================================================| 100%Examine the output of h2o.automl
models_h2o %>% typeof() ## [1] "S4"models_h2o %>% slotNames()## [1] "project_name" "leader" "leaderboard" "event_log"
## [5] "modeling_steps" "training_info"models_h2o@leaderboard## model_id mean_per_class_error logloss
## 1 GBM_3_AutoML_1_20241206_155258 0.2231392 0.3855270
## 2 GBM_5_AutoML_1_20241206_155258 0.2286947 0.3891497
## 3 GBM_2_AutoML_1_20241206_155258 0.2408846 0.3901091
## 4 GBM_grid_1_AutoML_1_20241206_155258_model_2 0.2432039 0.4039940
## 5 GBM_4_AutoML_1_20241206_155258 0.2586300 0.3972846
## 6 GBM_grid_1_AutoML_1_20241206_155258_model_3 0.2643474 0.3742388
## rmse mse
## 1 0.3275068 0.1072607
## 2 0.3297208 0.1087158
## 3 0.3294560 0.1085413
## 4 0.3305562 0.1092674
## 5 0.3325350 0.1105795
## 6 0.3300422 0.1089278
##
## [12 rows x 5 columns]models_h2o@leader## Model Details:
## ==============
##
## H2OMultinomialModel: gbm
## Model ID: GBM_3_AutoML_1_20241206_155258
## Model Summary:
## number_of_trees number_of_internal_trees model_size_in_bytes min_depth
## 1 40 120 65387 2
## max_depth mean_depth min_leaves max_leaves mean_leaves
## 1 8 7.95000 3 50 38.74166
##
##
## H2OMultinomialMetrics: gbm
## ** Reported on training data. **
##
## Training Set Metrics:
## =====================
##
## Extract training frame with `h2o.getFrame("AutoML_1_20241206_155258_training_RTMP_sid_82bd_5")`
## MSE: (Extract with `h2o.mse`) 0.008729674
## RMSE: (Extract with `h2o.rmse`) 0.09343272
## Logloss: (Extract with `h2o.logloss`) 0.06493089
## Mean Per-Class Error: 0.004444444
## AUC: (Extract with `h2o.auc`) NaN
## AUCPR: (Extract with `h2o.aucpr`) NaN
## R^2: (Extract with `h2o.r2`) 0.9356362
## Confusion Matrix: Extract with `h2o.confusionMatrix(<model>,train = TRUE)`)
## =========================================================================
## Confusion Matrix: Row labels: Actual class; Column labels: Predicted class
## Attrition No Yes Error Rate
## Attrition 1 0 0 0.0000 = 0 / 1
## No 0 788 0 0.0000 = 0 / 788
## Yes 0 2 148 0.0133 = 2 / 150
## Totals 1 790 148 0.0021 = 2 / 939
##
## Hit Ratio Table: Extract with `h2o.hit_ratio_table(<model>,train = TRUE)`
## =======================================================================
## Top-3 Hit Ratios:
## k hit_ratio
## 1 1 0.997870
## 2 2 1.000000
## 3 3 1.000000
##
##
##
##
## H2OMultinomialMetrics: gbm
## ** Reported on validation data. **
## ** Validation metrics **
##
## Validation Set Metrics:
## =====================
##
## Extract validation frame with `h2o.getFrame("RTMP_sid_82bd_7")`
## MSE: (Extract with `h2o.mse`) 0.101691
## RMSE: (Extract with `h2o.rmse`) 0.3188903
## Logloss: (Extract with `h2o.logloss`) 0.3544887
## Mean Per-Class Error: 0.2394699
## AUC: (Extract with `h2o.auc`) NaN
## AUCPR: (Extract with `h2o.aucpr`) NaN
## R^2: (Extract with `h2o.r2`) 0.264208
## Confusion Matrix: Extract with `h2o.confusionMatrix(<model>,valid = TRUE)`)
## =========================================================================
## Confusion Matrix: Row labels: Actual class; Column labels: Predicted class
## Attrition No Yes Error Rate
## Attrition 0 0 0 NA = 0 / 0
## No 0 134 2 0.0147 = 2 / 136
## Yes 0 19 8 0.7037 = 19 / 27
## Totals 0 153 10 0.1288 = 21 / 163
##
## Hit Ratio Table: Extract with `h2o.hit_ratio_table(<model>,valid = TRUE)`
## =======================================================================
## Top-3 Hit Ratios:
## k hit_ratio
## 1 1 0.871166
## 2 2 1.000000
## 3 3 1.000000
##
##
##
##
## H2OMultinomialMetrics: gbm
## ** Reported on cross-validation data. **
## ** 5-fold cross-validation on training data (Metrics computed for combined holdout predictions) **
##
## Cross-Validation Set Metrics:
## =====================
##
## Extract cross-validation frame with `h2o.getFrame("AutoML_1_20241206_155258_training_RTMP_sid_82bd_5")`
## MSE: (Extract with `h2o.mse`) 0.1092226
## RMSE: (Extract with `h2o.rmse`) 0.3304884
## Logloss: (Extract with `h2o.logloss`) 0.3815745
## Mean Per-Class Error: 0.5957642
## AUC: (Extract with `h2o.auc`) NaN
## AUCPR: (Extract with `h2o.aucpr`) NaN
## R^2: (Extract with `h2o.r2`) 0.194703
## Hit Ratio Table: Extract with `h2o.hit_ratio_table(<model>,xval = TRUE)`
## =======================================================================
## Top-3 Hit Ratios:
## k hit_ratio
## 1 1 0.863685
## 2 2 0.988285
## 3 3 1.000000
##
##
##
##
## Cross-Validation Metrics Summary:
## mean sd cv_1_valid cv_2_valid cv_3_valid
## accuracy 0.861583 0.015252 0.856383 0.851064 0.851064
## auc NA 0.000000 NA NA NA
## err 0.138417 0.015252 0.143617 0.148936 0.148936
## err_count 26.000000 2.915476 27.000000 28.000000 28.000000
## logloss 0.393179 0.053331 0.466507 0.387792 0.423399
## max_per_class_error 0.826667 0.136219 1.000000 0.800000 0.900000
## mean_per_class_accuracy 0.671852 0.142794 0.422718 0.724894 0.697890
## mean_per_class_error 0.328148 0.142794 0.577282 0.275105 0.302110
## mse 0.109744 0.014321 0.126312 0.112389 0.117847
## pr_auc NA 0.000000 NA NA NA
## r2 0.191591 0.095742 0.104797 0.161965 0.121266
## rmse 0.330694 0.021947 0.355404 0.335245 0.343289
## cv_4_valid cv_5_valid
## accuracy 0.861702 0.887701
## auc NA NA
## err 0.138298 0.112299
## err_count 26.000000 21.000000
## logloss 0.355043 0.333153
## max_per_class_error 0.800000 0.633333
## mean_per_class_accuracy 0.729114 0.784643
## mean_per_class_error 0.270886 0.215357
## mse 0.103161 0.089010
## pr_auc NA NA
## r2 0.230777 0.339151
## rmse 0.321187 0.298346best_model <- models_h2o@leader
best_model## Model Details:
## ==============
##
## H2OMultinomialModel: gbm
## Model ID: GBM_3_AutoML_1_20241206_155258
## Model Summary:
## number_of_trees number_of_internal_trees model_size_in_bytes min_depth
## 1 40 120 65387 2
## max_depth mean_depth min_leaves max_leaves mean_leaves
## 1 8 7.95000 3 50 38.74166
##
##
## H2OMultinomialMetrics: gbm
## ** Reported on training data. **
##
## Training Set Metrics:
## =====================
##
## Extract training frame with `h2o.getFrame("AutoML_1_20241206_155258_training_RTMP_sid_82bd_5")`
## MSE: (Extract with `h2o.mse`) 0.008729674
## RMSE: (Extract with `h2o.rmse`) 0.09343272
## Logloss: (Extract with `h2o.logloss`) 0.06493089
## Mean Per-Class Error: 0.004444444
## AUC: (Extract with `h2o.auc`) NaN
## AUCPR: (Extract with `h2o.aucpr`) NaN
## R^2: (Extract with `h2o.r2`) 0.9356362
## Confusion Matrix: Extract with `h2o.confusionMatrix(<model>,train = TRUE)`)
## =========================================================================
## Confusion Matrix: Row labels: Actual class; Column labels: Predicted class
## Attrition No Yes Error Rate
## Attrition 1 0 0 0.0000 = 0 / 1
## No 0 788 0 0.0000 = 0 / 788
## Yes 0 2 148 0.0133 = 2 / 150
## Totals 1 790 148 0.0021 = 2 / 939
##
## Hit Ratio Table: Extract with `h2o.hit_ratio_table(<model>,train = TRUE)`
## =======================================================================
## Top-3 Hit Ratios:
## k hit_ratio
## 1 1 0.997870
## 2 2 1.000000
## 3 3 1.000000
##
##
##
##
## H2OMultinomialMetrics: gbm
## ** Reported on validation data. **
## ** Validation metrics **
##
## Validation Set Metrics:
## =====================
##
## Extract validation frame with `h2o.getFrame("RTMP_sid_82bd_7")`
## MSE: (Extract with `h2o.mse`) 0.101691
## RMSE: (Extract with `h2o.rmse`) 0.3188903
## Logloss: (Extract with `h2o.logloss`) 0.3544887
## Mean Per-Class Error: 0.2394699
## AUC: (Extract with `h2o.auc`) NaN
## AUCPR: (Extract with `h2o.aucpr`) NaN
## R^2: (Extract with `h2o.r2`) 0.264208
## Confusion Matrix: Extract with `h2o.confusionMatrix(<model>,valid = TRUE)`)
## =========================================================================
## Confusion Matrix: Row labels: Actual class; Column labels: Predicted class
## Attrition No Yes Error Rate
## Attrition 0 0 0 NA = 0 / 0
## No 0 134 2 0.0147 = 2 / 136
## Yes 0 19 8 0.7037 = 19 / 27
## Totals 0 153 10 0.1288 = 21 / 163
##
## Hit Ratio Table: Extract with `h2o.hit_ratio_table(<model>,valid = TRUE)`
## =======================================================================
## Top-3 Hit Ratios:
## k hit_ratio
## 1 1 0.871166
## 2 2 1.000000
## 3 3 1.000000
##
##
##
##
## H2OMultinomialMetrics: gbm
## ** Reported on cross-validation data. **
## ** 5-fold cross-validation on training data (Metrics computed for combined holdout predictions) **
##
## Cross-Validation Set Metrics:
## =====================
##
## Extract cross-validation frame with `h2o.getFrame("AutoML_1_20241206_155258_training_RTMP_sid_82bd_5")`
## MSE: (Extract with `h2o.mse`) 0.1092226
## RMSE: (Extract with `h2o.rmse`) 0.3304884
## Logloss: (Extract with `h2o.logloss`) 0.3815745
## Mean Per-Class Error: 0.5957642
## AUC: (Extract with `h2o.auc`) NaN
## AUCPR: (Extract with `h2o.aucpr`) NaN
## R^2: (Extract with `h2o.r2`) 0.194703
## Hit Ratio Table: Extract with `h2o.hit_ratio_table(<model>,xval = TRUE)`
## =======================================================================
## Top-3 Hit Ratios:
## k hit_ratio
## 1 1 0.863685
## 2 2 0.988285
## 3 3 1.000000
##
##
##
##
## Cross-Validation Metrics Summary:
## mean sd cv_1_valid cv_2_valid cv_3_valid
## accuracy 0.861583 0.015252 0.856383 0.851064 0.851064
## auc NA 0.000000 NA NA NA
## err 0.138417 0.015252 0.143617 0.148936 0.148936
## err_count 26.000000 2.915476 27.000000 28.000000 28.000000
## logloss 0.393179 0.053331 0.466507 0.387792 0.423399
## max_per_class_error 0.826667 0.136219 1.000000 0.800000 0.900000
## mean_per_class_accuracy 0.671852 0.142794 0.422718 0.724894 0.697890
## mean_per_class_error 0.328148 0.142794 0.577282 0.275105 0.302110
## mse 0.109744 0.014321 0.126312 0.112389 0.117847
## pr_auc NA 0.000000 NA NA NA
## r2 0.191591 0.095742 0.104797 0.161965 0.121266
## rmse 0.330694 0.021947 0.355404 0.335245 0.343289
## cv_4_valid cv_5_valid
## accuracy 0.861702 0.887701
## auc NA NA
## err 0.138298 0.112299
## err_count 26.000000 21.000000
## logloss 0.355043 0.333153
## max_per_class_error 0.800000 0.633333
## mean_per_class_accuracy 0.729114 0.784643
## mean_per_class_error 0.270886 0.215357
## mse 0.103161 0.089010
## pr_auc NA NA
## r2 0.230777 0.339151
## rmse 0.321187 0.298346#?h2o.getModel
#?h2o.saveModel
#?h2o.loadModel
#h2o.getModel(GBM_3_AutoML_1_20241121_121503) %>%
# h2o.saveModel("h2o_models/")
#best_model <- h2o.loadModel(GBM_3_AutoML_1_20241121_121503)predictions <- h2o.predict(best_model, newdata = test_h2o)##
|
| | 0%
|
|======================================================================| 100%predictions_tbl <- predictions %>%
as_tibble()
# predictions_tbl %>%
# bind_cols(test_tbl)?h2o.performance## starting httpd help server ... doneperformance_h2o <- h2o.performance(best_model, newdata = as.h2o(test_tbl))##
|
| | 0%
|
|======================================================================| 100%typeof(performance_h2o)## [1] "S4"slotNames(performance_h2o)## [1] "algorithm" "on_train" "on_valid" "on_xval" "metrics"performance_h2o@metrics## $model
## $model$`__meta`
## $model$`__meta`$schema_version
## [1] 3
##
## $model$`__meta`$schema_name
## [1] "ModelKeyV3"
##
## $model$`__meta`$schema_type
## [1] "Key<Model>"
##
##
## $model$name
## [1] "GBM_3_AutoML_1_20241206_155258"
##
## $model$type
## [1] "Key<Model>"
##
## $model$URL
## [1] "/3/Models/GBM_3_AutoML_1_20241206_155258"
##
##
## $model_checksum
## [1] "3271101342999854048"
##
## $frame
## $frame$name
## [1] "test_tbl_sid_82bd_119"
##
##
## $frame_checksum
## [1] "-5977174161532928616"
##
## $description
## NULL
##
## $scoring_time
## [1] 1.733519e+12
##
## $predictions
## NULL
##
## $MSE
## [1] 0.1072607
##
## $RMSE
## [1] 0.3275068
##
## $nobs
## [1] 370
##
## $custom_metric_name
## NULL
##
## $custom_metric_value
## [1] 0
##
## $r2
## [1] 0.2307616
##
## $hit_ratio_table
## Top-3 Hit Ratios:
## k hit_ratio
## 1 1 0.878378
## 2 2 1.000000
## 3 3 1.000000
##
## $cm
## $cm$`__meta`
## $cm$`__meta`$schema_version
## [1] 3
##
## $cm$`__meta`$schema_name
## [1] "ConfusionMatrixV3"
##
## $cm$`__meta`$schema_type
## [1] "ConfusionMatrix"
##
##
## $cm$table
## Confusion Matrix: Row labels: Actual class; Column labels: Predicted class
## Attrition No Yes Error Rate
## Attrition 1 0 0 0.0000 = 0 / 1
## No 0 303 6 0.0194 = 6 / 309
## Yes 0 39 21 0.6500 = 39 / 60
## Totals 1 342 27 0.1216 = 45 / 370
##
##
## $logloss
## [1] 0.385527
##
## $mean_per_class_error
## [1] 0.2231392
##
## $AUC
## [1] "NaN"
##
## $pr_auc
## [1] "NaN"
##
## $multinomial_auc_table
## NULL
##
## $multinomial_aucpr_table
## NULLh2o.auc(performance_h2o)## [1] "NaN"h2o.confusionMatrix(performance_h2o)## Confusion Matrix: Row labels: Actual class; Column labels: Predicted class
## Attrition No Yes Error Rate
## Attrition 1 0 0 0.0000 = 0 / 1
## No 0 303 6 0.0194 = 6 / 309
## Yes 0 39 21 0.6500 = 39 / 60
## Totals 1 342 27 0.1216 = 45 / 370