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 4.2.3##
## ----------------------------------------------------------------------
##
## 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 'tidyverse' was built under R version 4.2.3## Warning: package 'ggplot2' was built under R version 4.2.3## Warning: package 'tibble' was built under R version 4.2.3## Warning: package 'tidyr' was built under R version 4.2.3## Warning: package 'readr' was built under R version 4.2.3## Warning: package 'purrr' was built under R version 4.2.3## Warning: package 'dplyr' was built under R version 4.2.3## Warning: package 'stringr' was built under R version 4.2.3## Warning: package 'forcats' was built under R version 4.2.3## Warning: package 'lubridate' was built under R version 4.2.3## ── 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 3.4.4 ✔ tibble 3.2.1
## ✔ lubridate 1.9.2 ✔ 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 4.2.3## ── Attaching packages ────────────────────────────────────── tidymodels 1.1.1 ──
## ✔ broom 1.0.5 ✔ rsample 1.2.0
## ✔ dials 1.2.0 ✔ tune 1.1.2
## ✔ infer 1.0.6 ✔ workflows 1.1.3
## ✔ modeldata 1.3.0 ✔ workflowsets 1.0.1
## ✔ parsnip 1.1.1 ✔ yardstick 1.3.0
## ✔ recipes 1.0.9## Warning: package 'broom' was built under R version 4.2.3## Warning: package 'dials' was built under R version 4.2.3## Warning: package 'infer' was built under R version 4.2.3## Warning: package 'modeldata' was built under R version 4.2.3## Warning: package 'parsnip' was built under R version 4.2.3## Warning: package 'recipes' was built under R version 4.2.3## Warning: package 'rsample' was built under R version 4.2.3## Warning: package 'tune' was built under R version 4.2.3## Warning: package 'workflows' was built under R version 4.2.3## Warning: package 'workflowsets' was built under R version 4.2.3## Warning: package 'yardstick' was built under R version 4.2.3## ── 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()
## • Search for functions across packages at https://www.tidymodels.org/find/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
##
##
## 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: 32
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (8): Attrition, BusinessTravel, Department, EducationField, Gender, Job...
## dbl (24): Age, DailyRate, DistanceFromHome, Education, EmployeeNumber, Envir...
##
## 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()## Connection successful!
##
## R is connected to the H2O cluster:
## H2O cluster uptime: 33 minutes 433 milliseconds
## 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 3 days
## H2O cluster name: H2O_started_from_R_aesim_fhp551
## H2O cluster total nodes: 1
## H2O cluster total memory: 3.31 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 4.2.2 (2022-10-31 ucrt)## Warning in h2o.clusterInfo():
## Your H2O cluster version is (4 months and 3 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), ratios = c(0.85), seed = 2567)##
|
| | 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)
auto_ml_models_h2o <- h2o.automl(
x = x,
y = y,
training_frame = train_h2o,
validation_frame = valid_h2o,
leaderboard_frame = test_h2o,
max_runtime_secs = 30,
nfolds = 5,
seed = 3456)##
|
| | 0%
|
|==== | 6%
## 18:48:56.927: 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.
## 18:48:56.931: AutoML: XGBoost is not available; skipping it.
|
|============= | 18%
|
|==================== | 28%
|
|====================== | 31%
|
|============================ | 40%
|
|=================================== | 49%
|
|================================================== | 72%
|
|========================================================== | 82%
|
|================================================================ | 91%
|
|======================================================================| 100%auto_ml_models_h2o@leaderboard## model_id auc logloss
## 1 DeepLearning_grid_1_AutoML_7_20240423_184856_model_1 0.8673139 0.3784866
## 2 StackedEnsemble_BestOfFamily_4_AutoML_7_20240423_184856 0.8640237 0.3079303
## 3 GBM_grid_1_AutoML_7_20240423_184856_model_14 0.8528587 0.3232253
## 4 StackedEnsemble_BestOfFamily_3_AutoML_7_20240423_184856 0.8515642 0.3030396
## 5 StackedEnsemble_BestOfFamily_2_AutoML_7_20240423_184856 0.8515642 0.3030082
## 6 StackedEnsemble_BestOfFamily_1_AutoML_7_20240423_184856 0.8513484 0.3028647
## aucpr mean_per_class_error rmse mse
## 1 0.6842232 0.2307443 0.3144402 0.09887263
## 2 0.6915659 0.1920712 0.3018720 0.09112671
## 3 0.5962635 0.2474110 0.3120479 0.09737391
## 4 0.6703203 0.2004045 0.2948439 0.08693295
## 5 0.6703203 0.2004045 0.2948577 0.08694106
## 6 0.6696650 0.2004045 0.2949408 0.08699008
##
## [40 rows x 7 columns]auto_ml_models_h2o@leader## Model Details:
## ==============
##
## H2OBinomialModel: deeplearning
## Model ID: DeepLearning_grid_1_AutoML_7_20240423_184856_model_1
## Status of Neuron Layers: predicting Attrition, 2-class classification, bernoulli distribution, CrossEntropy loss, 6,202 weights/biases, 83.3 KB, 8,055 training samples, mini-batch size 1
## layer units type dropout l1 l2 mean_rate rate_rms
## 1 1 59 Input 5.00 % NA NA NA NA
## 2 2 100 RectifierDropout 20.00 % 0.000000 0.000000 0.118115 0.316928
## 3 3 2 Softmax NA 0.000000 0.000000 0.000426 0.000089
## momentum mean_weight weight_rms mean_bias bias_rms
## 1 NA NA NA NA NA
## 2 0.000000 0.000493 0.110705 0.487810 0.020024
## 3 0.000000 -0.043842 0.559034 -0.000409 0.010533
##
##
## H2OBinomialMetrics: deeplearning
## ** Reported on training data. **
## ** Metrics reported on full training frame **
##
## MSE: 0.0885245
## RMSE: 0.2975307
## LogLoss: 0.3354281
## Mean Per-Class Error: 0.1882202
## AUC: 0.8815982
## AUCPR: 0.7056277
## Gini: 0.7631963
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## No Yes Error Rate
## No 729 52 0.066581 =52/781
## Yes 44 98 0.309859 =44/142
## Totals 773 150 0.104009 =96/923
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.118543 0.671233 142
## 2 max f2 0.043056 0.696822 216
## 3 max f0point5 0.204088 0.698925 100
## 4 max accuracy 0.204088 0.902492 100
## 5 max precision 0.993111 1.000000 0
## 6 max recall 0.000024 1.000000 399
## 7 max specificity 0.993111 1.000000 0
## 8 max absolute_mcc 0.141067 0.610621 128
## 9 max min_per_class_accuracy 0.043056 0.802817 216
## 10 max mean_per_class_accuracy 0.087788 0.819462 155
## 11 max tns 0.993111 781.000000 0
## 12 max fns 0.993111 141.000000 0
## 13 max fps 0.000024 781.000000 399
## 14 max tps 0.000024 142.000000 399
## 15 max tnr 0.993111 1.000000 0
## 16 max fnr 0.993111 0.992958 0
## 17 max fpr 0.000024 1.000000 399
## 18 max tpr 0.000024 1.000000 399
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## H2OBinomialMetrics: deeplearning
## ** Reported on validation data. **
## ** Metrics reported on full validation frame **
##
## MSE: 0.1333056
## RMSE: 0.3651103
## LogLoss: 0.5772558
## Mean Per-Class Error: 0.281019
## AUC: 0.784016
## AUCPR: 0.5980084
## Gini: 0.568032
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## No Yes Error Rate
## No 128 15 0.104895 =15/143
## Yes 16 19 0.457143 =16/35
## Totals 144 34 0.174157 =31/178
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.073982 0.550725 33
## 2 max f2 0.005891 0.648536 98
## 3 max f0point5 0.293960 0.659341 13
## 4 max accuracy 0.293960 0.859551 13
## 5 max precision 0.970038 1.000000 0
## 6 max recall 0.000438 1.000000 157
## 7 max specificity 0.970038 1.000000 0
## 8 max absolute_mcc 0.293960 0.485559 13
## 9 max min_per_class_accuracy 0.019448 0.685714 65
## 10 max mean_per_class_accuracy 0.073982 0.718981 33
## 11 max tns 0.970038 143.000000 0
## 12 max fns 0.970038 34.000000 0
## 13 max fps 0.000003 143.000000 177
## 14 max tps 0.000438 35.000000 157
## 15 max tnr 0.970038 1.000000 0
## 16 max fnr 0.970038 0.971429 0
## 17 max fpr 0.000003 1.000000 177
## 18 max tpr 0.000438 1.000000 157
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## H2OBinomialMetrics: deeplearning
## ** Reported on cross-validation data. **
## ** 5-fold cross-validation on training data (Metrics computed for combined holdout predictions) **
##
## MSE: 0.1434243
## RMSE: 0.378714
## LogLoss: 0.9498462
## Mean Per-Class Error: 0.2291933
## AUC: 0.8277714
## AUCPR: 0.5620906
## Gini: 0.6555427
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## No Yes Error Rate
## No 698 83 0.106274 =83/781
## Yes 50 92 0.352113 =50/142
## Totals 748 175 0.144095 =133/923
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.001488 0.580442 170
## 2 max f2 0.000387 0.634638 252
## 3 max f0point5 0.003443 0.559105 118
## 4 max accuracy 0.017843 0.874323 43
## 5 max precision 0.846356 1.000000 0
## 6 max recall 0.000000 1.000000 399
## 7 max specificity 0.846356 1.000000 0
## 8 max absolute_mcc 0.001488 0.498528 170
## 9 max min_per_class_accuracy 0.000387 0.753521 252
## 10 max mean_per_class_accuracy 0.001488 0.770807 170
## 11 max tns 0.846356 781.000000 0
## 12 max fns 0.846356 141.000000 0
## 13 max fps 0.000000 781.000000 399
## 14 max tps 0.000000 142.000000 399
## 15 max tnr 0.846356 1.000000 0
## 16 max fnr 0.846356 0.992958 0
## 17 max fpr 0.000000 1.000000 399
## 18 max tpr 0.000000 1.000000 399
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## Cross-Validation Metrics Summary:
## mean sd cv_1_valid cv_2_valid cv_3_valid
## accuracy 0.814841 0.071526 0.702703 0.848649 0.794595
## auc 0.790328 0.071853 0.666060 0.809682 0.803050
## err 0.185159 0.071526 0.297297 0.151351 0.205405
## err_count 34.200000 13.274035 55.000000 28.000000 38.000000
## f0point5 0.484620 0.124550 0.303030 0.522876 0.450644
## f1 0.514321 0.086962 0.367816 0.533333 0.525000
## f2 0.560100 0.061378 0.467836 0.544218 0.628742
## lift_top_group 5.863793 1.497568 6.607143 3.189655 6.379310
## logloss 1.185119 0.174568 1.261851 1.237098 1.387543
## max_per_class_error 0.394828 0.078091 0.428571 0.448276 0.275862
## mcc 0.421009 0.116009 0.228806 0.443481 0.432710
## mean_per_class_accuracy 0.729044 0.047098 0.648772 0.727785 0.765915
## mean_per_class_error 0.270956 0.047098 0.351228 0.272215 0.234085
## mse 0.150494 0.006590 0.149274 0.156200 0.156519
## pr_auc 0.504576 0.107795 0.320343 0.499387 0.560233
## precision 0.470917 0.149935 0.271186 0.516129 0.411765
## r2 -0.155887 0.039403 -0.162174 -0.181687 -0.184098
## recall 0.605172 0.078091 0.571429 0.551724 0.724138
## rmse 0.387861 0.008550 0.386360 0.395222 0.395625
## specificity 0.852915 0.088771 0.726115 0.903846 0.807692
## cv_4_valid cv_5_valid
## accuracy 0.891304 0.836956
## auc 0.850962 0.821886
## err 0.108696 0.163043
## err_count 20.000000 30.000000
## f0point5 0.646552 0.500000
## f1 0.600000 0.545455
## f2 0.559701 0.600000
## lift_top_group 6.571429 6.571429
## logloss 1.113108 0.925995
## max_per_class_error 0.464286 0.357143
## mcc 0.543394 0.456653
## mean_per_class_accuracy 0.745421 0.757326
## mean_per_class_error 0.254579 0.242674
## mse 0.150145 0.140334
## pr_auc 0.586724 0.556194
## precision 0.681818 0.473684
## r2 -0.163760 -0.087715
## recall 0.535714 0.642857
## rmse 0.387485 0.374611
## specificity 0.955128 0.871795best_model <- auto_ml_models_h2o@leader
best_model## Model Details:
## ==============
##
## H2OBinomialModel: deeplearning
## Model ID: DeepLearning_grid_1_AutoML_7_20240423_184856_model_1
## Status of Neuron Layers: predicting Attrition, 2-class classification, bernoulli distribution, CrossEntropy loss, 6,202 weights/biases, 83.3 KB, 8,055 training samples, mini-batch size 1
## layer units type dropout l1 l2 mean_rate rate_rms
## 1 1 59 Input 5.00 % NA NA NA NA
## 2 2 100 RectifierDropout 20.00 % 0.000000 0.000000 0.118115 0.316928
## 3 3 2 Softmax NA 0.000000 0.000000 0.000426 0.000089
## momentum mean_weight weight_rms mean_bias bias_rms
## 1 NA NA NA NA NA
## 2 0.000000 0.000493 0.110705 0.487810 0.020024
## 3 0.000000 -0.043842 0.559034 -0.000409 0.010533
##
##
## H2OBinomialMetrics: deeplearning
## ** Reported on training data. **
## ** Metrics reported on full training frame **
##
## MSE: 0.0885245
## RMSE: 0.2975307
## LogLoss: 0.3354281
## Mean Per-Class Error: 0.1882202
## AUC: 0.8815982
## AUCPR: 0.7056277
## Gini: 0.7631963
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## No Yes Error Rate
## No 729 52 0.066581 =52/781
## Yes 44 98 0.309859 =44/142
## Totals 773 150 0.104009 =96/923
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.118543 0.671233 142
## 2 max f2 0.043056 0.696822 216
## 3 max f0point5 0.204088 0.698925 100
## 4 max accuracy 0.204088 0.902492 100
## 5 max precision 0.993111 1.000000 0
## 6 max recall 0.000024 1.000000 399
## 7 max specificity 0.993111 1.000000 0
## 8 max absolute_mcc 0.141067 0.610621 128
## 9 max min_per_class_accuracy 0.043056 0.802817 216
## 10 max mean_per_class_accuracy 0.087788 0.819462 155
## 11 max tns 0.993111 781.000000 0
## 12 max fns 0.993111 141.000000 0
## 13 max fps 0.000024 781.000000 399
## 14 max tps 0.000024 142.000000 399
## 15 max tnr 0.993111 1.000000 0
## 16 max fnr 0.993111 0.992958 0
## 17 max fpr 0.000024 1.000000 399
## 18 max tpr 0.000024 1.000000 399
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## H2OBinomialMetrics: deeplearning
## ** Reported on validation data. **
## ** Metrics reported on full validation frame **
##
## MSE: 0.1333056
## RMSE: 0.3651103
## LogLoss: 0.5772558
## Mean Per-Class Error: 0.281019
## AUC: 0.784016
## AUCPR: 0.5980084
## Gini: 0.568032
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## No Yes Error Rate
## No 128 15 0.104895 =15/143
## Yes 16 19 0.457143 =16/35
## Totals 144 34 0.174157 =31/178
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.073982 0.550725 33
## 2 max f2 0.005891 0.648536 98
## 3 max f0point5 0.293960 0.659341 13
## 4 max accuracy 0.293960 0.859551 13
## 5 max precision 0.970038 1.000000 0
## 6 max recall 0.000438 1.000000 157
## 7 max specificity 0.970038 1.000000 0
## 8 max absolute_mcc 0.293960 0.485559 13
## 9 max min_per_class_accuracy 0.019448 0.685714 65
## 10 max mean_per_class_accuracy 0.073982 0.718981 33
## 11 max tns 0.970038 143.000000 0
## 12 max fns 0.970038 34.000000 0
## 13 max fps 0.000003 143.000000 177
## 14 max tps 0.000438 35.000000 157
## 15 max tnr 0.970038 1.000000 0
## 16 max fnr 0.970038 0.971429 0
## 17 max fpr 0.000003 1.000000 177
## 18 max tpr 0.000438 1.000000 157
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## H2OBinomialMetrics: deeplearning
## ** Reported on cross-validation data. **
## ** 5-fold cross-validation on training data (Metrics computed for combined holdout predictions) **
##
## MSE: 0.1434243
## RMSE: 0.378714
## LogLoss: 0.9498462
## Mean Per-Class Error: 0.2291933
## AUC: 0.8277714
## AUCPR: 0.5620906
## Gini: 0.6555427
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## No Yes Error Rate
## No 698 83 0.106274 =83/781
## Yes 50 92 0.352113 =50/142
## Totals 748 175 0.144095 =133/923
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.001488 0.580442 170
## 2 max f2 0.000387 0.634638 252
## 3 max f0point5 0.003443 0.559105 118
## 4 max accuracy 0.017843 0.874323 43
## 5 max precision 0.846356 1.000000 0
## 6 max recall 0.000000 1.000000 399
## 7 max specificity 0.846356 1.000000 0
## 8 max absolute_mcc 0.001488 0.498528 170
## 9 max min_per_class_accuracy 0.000387 0.753521 252
## 10 max mean_per_class_accuracy 0.001488 0.770807 170
## 11 max tns 0.846356 781.000000 0
## 12 max fns 0.846356 141.000000 0
## 13 max fps 0.000000 781.000000 399
## 14 max tps 0.000000 142.000000 399
## 15 max tnr 0.846356 1.000000 0
## 16 max fnr 0.846356 0.992958 0
## 17 max fpr 0.000000 1.000000 399
## 18 max tpr 0.000000 1.000000 399
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## Cross-Validation Metrics Summary:
## mean sd cv_1_valid cv_2_valid cv_3_valid
## accuracy 0.814841 0.071526 0.702703 0.848649 0.794595
## auc 0.790328 0.071853 0.666060 0.809682 0.803050
## err 0.185159 0.071526 0.297297 0.151351 0.205405
## err_count 34.200000 13.274035 55.000000 28.000000 38.000000
## f0point5 0.484620 0.124550 0.303030 0.522876 0.450644
## f1 0.514321 0.086962 0.367816 0.533333 0.525000
## f2 0.560100 0.061378 0.467836 0.544218 0.628742
## lift_top_group 5.863793 1.497568 6.607143 3.189655 6.379310
## logloss 1.185119 0.174568 1.261851 1.237098 1.387543
## max_per_class_error 0.394828 0.078091 0.428571 0.448276 0.275862
## mcc 0.421009 0.116009 0.228806 0.443481 0.432710
## mean_per_class_accuracy 0.729044 0.047098 0.648772 0.727785 0.765915
## mean_per_class_error 0.270956 0.047098 0.351228 0.272215 0.234085
## mse 0.150494 0.006590 0.149274 0.156200 0.156519
## pr_auc 0.504576 0.107795 0.320343 0.499387 0.560233
## precision 0.470917 0.149935 0.271186 0.516129 0.411765
## r2 -0.155887 0.039403 -0.162174 -0.181687 -0.184098
## recall 0.605172 0.078091 0.571429 0.551724 0.724138
## rmse 0.387861 0.008550 0.386360 0.395222 0.395625
## specificity 0.852915 0.088771 0.726115 0.903846 0.807692
## cv_4_valid cv_5_valid
## accuracy 0.891304 0.836956
## auc 0.850962 0.821886
## err 0.108696 0.163043
## err_count 20.000000 30.000000
## f0point5 0.646552 0.500000
## f1 0.600000 0.545455
## f2 0.559701 0.600000
## lift_top_group 6.571429 6.571429
## logloss 1.113108 0.925995
## max_per_class_error 0.464286 0.357143
## mcc 0.543394 0.456653
## mean_per_class_accuracy 0.745421 0.757326
## mean_per_class_error 0.254579 0.242674
## mse 0.150145 0.140334
## pr_auc 0.586724 0.556194
## precision 0.681818 0.473684
## r2 -0.163760 -0.087715
## recall 0.535714 0.642857
## rmse 0.387485 0.374611
## specificity 0.955128 0.871795Examine the output of h2o.automl
auto_ml_models_h2o %>% typeof()## [1] "S4"auto_ml_models_h2o %>% slotNames()## [1] "project_name" "leader" "leaderboard" "event_log"
## [5] "modeling_steps" "training_info"auto_ml_models_h2o@leaderboard## model_id auc logloss
## 1 DeepLearning_grid_1_AutoML_7_20240423_184856_model_1 0.8673139 0.3784866
## 2 StackedEnsemble_BestOfFamily_4_AutoML_7_20240423_184856 0.8640237 0.3079303
## 3 GBM_grid_1_AutoML_7_20240423_184856_model_14 0.8528587 0.3232253
## 4 StackedEnsemble_BestOfFamily_3_AutoML_7_20240423_184856 0.8515642 0.3030396
## 5 StackedEnsemble_BestOfFamily_2_AutoML_7_20240423_184856 0.8515642 0.3030082
## 6 StackedEnsemble_BestOfFamily_1_AutoML_7_20240423_184856 0.8513484 0.3028647
## aucpr mean_per_class_error rmse mse
## 1 0.6842232 0.2307443 0.3144402 0.09887263
## 2 0.6915659 0.1920712 0.3018720 0.09112671
## 3 0.5962635 0.2474110 0.3120479 0.09737391
## 4 0.6703203 0.2004045 0.2948439 0.08693295
## 5 0.6703203 0.2004045 0.2948577 0.08694106
## 6 0.6696650 0.2004045 0.2949408 0.08699008
##
## [40 rows x 7 columns]auto_ml_models_h2o@leader## Model Details:
## ==============
##
## H2OBinomialModel: deeplearning
## Model ID: DeepLearning_grid_1_AutoML_7_20240423_184856_model_1
## Status of Neuron Layers: predicting Attrition, 2-class classification, bernoulli distribution, CrossEntropy loss, 6,202 weights/biases, 83.3 KB, 8,055 training samples, mini-batch size 1
## layer units type dropout l1 l2 mean_rate rate_rms
## 1 1 59 Input 5.00 % NA NA NA NA
## 2 2 100 RectifierDropout 20.00 % 0.000000 0.000000 0.118115 0.316928
## 3 3 2 Softmax NA 0.000000 0.000000 0.000426 0.000089
## momentum mean_weight weight_rms mean_bias bias_rms
## 1 NA NA NA NA NA
## 2 0.000000 0.000493 0.110705 0.487810 0.020024
## 3 0.000000 -0.043842 0.559034 -0.000409 0.010533
##
##
## H2OBinomialMetrics: deeplearning
## ** Reported on training data. **
## ** Metrics reported on full training frame **
##
## MSE: 0.0885245
## RMSE: 0.2975307
## LogLoss: 0.3354281
## Mean Per-Class Error: 0.1882202
## AUC: 0.8815982
## AUCPR: 0.7056277
## Gini: 0.7631963
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## No Yes Error Rate
## No 729 52 0.066581 =52/781
## Yes 44 98 0.309859 =44/142
## Totals 773 150 0.104009 =96/923
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.118543 0.671233 142
## 2 max f2 0.043056 0.696822 216
## 3 max f0point5 0.204088 0.698925 100
## 4 max accuracy 0.204088 0.902492 100
## 5 max precision 0.993111 1.000000 0
## 6 max recall 0.000024 1.000000 399
## 7 max specificity 0.993111 1.000000 0
## 8 max absolute_mcc 0.141067 0.610621 128
## 9 max min_per_class_accuracy 0.043056 0.802817 216
## 10 max mean_per_class_accuracy 0.087788 0.819462 155
## 11 max tns 0.993111 781.000000 0
## 12 max fns 0.993111 141.000000 0
## 13 max fps 0.000024 781.000000 399
## 14 max tps 0.000024 142.000000 399
## 15 max tnr 0.993111 1.000000 0
## 16 max fnr 0.993111 0.992958 0
## 17 max fpr 0.000024 1.000000 399
## 18 max tpr 0.000024 1.000000 399
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## H2OBinomialMetrics: deeplearning
## ** Reported on validation data. **
## ** Metrics reported on full validation frame **
##
## MSE: 0.1333056
## RMSE: 0.3651103
## LogLoss: 0.5772558
## Mean Per-Class Error: 0.281019
## AUC: 0.784016
## AUCPR: 0.5980084
## Gini: 0.568032
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## No Yes Error Rate
## No 128 15 0.104895 =15/143
## Yes 16 19 0.457143 =16/35
## Totals 144 34 0.174157 =31/178
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.073982 0.550725 33
## 2 max f2 0.005891 0.648536 98
## 3 max f0point5 0.293960 0.659341 13
## 4 max accuracy 0.293960 0.859551 13
## 5 max precision 0.970038 1.000000 0
## 6 max recall 0.000438 1.000000 157
## 7 max specificity 0.970038 1.000000 0
## 8 max absolute_mcc 0.293960 0.485559 13
## 9 max min_per_class_accuracy 0.019448 0.685714 65
## 10 max mean_per_class_accuracy 0.073982 0.718981 33
## 11 max tns 0.970038 143.000000 0
## 12 max fns 0.970038 34.000000 0
## 13 max fps 0.000003 143.000000 177
## 14 max tps 0.000438 35.000000 157
## 15 max tnr 0.970038 1.000000 0
## 16 max fnr 0.970038 0.971429 0
## 17 max fpr 0.000003 1.000000 177
## 18 max tpr 0.000438 1.000000 157
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## H2OBinomialMetrics: deeplearning
## ** Reported on cross-validation data. **
## ** 5-fold cross-validation on training data (Metrics computed for combined holdout predictions) **
##
## MSE: 0.1434243
## RMSE: 0.378714
## LogLoss: 0.9498462
## Mean Per-Class Error: 0.2291933
## AUC: 0.8277714
## AUCPR: 0.5620906
## Gini: 0.6555427
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## No Yes Error Rate
## No 698 83 0.106274 =83/781
## Yes 50 92 0.352113 =50/142
## Totals 748 175 0.144095 =133/923
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.001488 0.580442 170
## 2 max f2 0.000387 0.634638 252
## 3 max f0point5 0.003443 0.559105 118
## 4 max accuracy 0.017843 0.874323 43
## 5 max precision 0.846356 1.000000 0
## 6 max recall 0.000000 1.000000 399
## 7 max specificity 0.846356 1.000000 0
## 8 max absolute_mcc 0.001488 0.498528 170
## 9 max min_per_class_accuracy 0.000387 0.753521 252
## 10 max mean_per_class_accuracy 0.001488 0.770807 170
## 11 max tns 0.846356 781.000000 0
## 12 max fns 0.846356 141.000000 0
## 13 max fps 0.000000 781.000000 399
## 14 max tps 0.000000 142.000000 399
## 15 max tnr 0.846356 1.000000 0
## 16 max fnr 0.846356 0.992958 0
## 17 max fpr 0.000000 1.000000 399
## 18 max tpr 0.000000 1.000000 399
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## Cross-Validation Metrics Summary:
## mean sd cv_1_valid cv_2_valid cv_3_valid
## accuracy 0.814841 0.071526 0.702703 0.848649 0.794595
## auc 0.790328 0.071853 0.666060 0.809682 0.803050
## err 0.185159 0.071526 0.297297 0.151351 0.205405
## err_count 34.200000 13.274035 55.000000 28.000000 38.000000
## f0point5 0.484620 0.124550 0.303030 0.522876 0.450644
## f1 0.514321 0.086962 0.367816 0.533333 0.525000
## f2 0.560100 0.061378 0.467836 0.544218 0.628742
## lift_top_group 5.863793 1.497568 6.607143 3.189655 6.379310
## logloss 1.185119 0.174568 1.261851 1.237098 1.387543
## max_per_class_error 0.394828 0.078091 0.428571 0.448276 0.275862
## mcc 0.421009 0.116009 0.228806 0.443481 0.432710
## mean_per_class_accuracy 0.729044 0.047098 0.648772 0.727785 0.765915
## mean_per_class_error 0.270956 0.047098 0.351228 0.272215 0.234085
## mse 0.150494 0.006590 0.149274 0.156200 0.156519
## pr_auc 0.504576 0.107795 0.320343 0.499387 0.560233
## precision 0.470917 0.149935 0.271186 0.516129 0.411765
## r2 -0.155887 0.039403 -0.162174 -0.181687 -0.184098
## recall 0.605172 0.078091 0.571429 0.551724 0.724138
## rmse 0.387861 0.008550 0.386360 0.395222 0.395625
## specificity 0.852915 0.088771 0.726115 0.903846 0.807692
## cv_4_valid cv_5_valid
## accuracy 0.891304 0.836956
## auc 0.850962 0.821886
## err 0.108696 0.163043
## err_count 20.000000 30.000000
## f0point5 0.646552 0.500000
## f1 0.600000 0.545455
## f2 0.559701 0.600000
## lift_top_group 6.571429 6.571429
## logloss 1.113108 0.925995
## max_per_class_error 0.464286 0.357143
## mcc 0.543394 0.456653
## mean_per_class_accuracy 0.745421 0.757326
## mean_per_class_error 0.254579 0.242674
## mse 0.150145 0.140334
## pr_auc 0.586724 0.556194
## precision 0.681818 0.473684
## r2 -0.163760 -0.087715
## recall 0.535714 0.642857
## rmse 0.387485 0.374611
## specificity 0.955128 0.871795# h2o.getModel("GBM_grid_1_AutoML_2_20240423_113107_model_1") %>%
# h2o.saveModel("h2o_models/")predictions <- h2o.predict(best_model, newdata = test_h2o)##
|
| | 0%
|
|======================================================================| 100%predictions_tbl <- predictions %>%
as_tibble()
predictions_tbl %>%
bind_cols(test_tbl)## # A tibble: 369 × 35
## predict No Yes Age Attrition BusinessTravel DailyRate Department
## <fct> <dbl> <dbl> <dbl> <fct> <fct> <dbl> <fct>
## 1 No 0.999 0.00125 59 No Travel_Rarely 1324 Research …
## 2 No 0.993 0.00721 35 No Travel_Rarely 809 Research …
## 3 No 0.993 0.00653 34 No Travel_Rarely 1346 Research …
## 4 No 0.985 0.0149 22 No Non-Travel 1123 Research …
## 5 No 1.00 0.000164 53 No Travel_Rarely 1219 Sales
## 6 No 0.983 0.0173 24 No Non-Travel 673 Research …
## 7 No 0.967 0.0334 21 No Travel_Rarely 391 Research …
## 8 No 0.977 0.0229 34 Yes Travel_Rarely 699 Research …
## 9 No 1.00 0.000340 53 No Travel_Rarely 1282 Research …
## 10 Yes 0.190 0.810 32 Yes Travel_Frequently 1125 Research …
## # ℹ 359 more rows
## # ℹ 27 more variables: DistanceFromHome <dbl>, Education <dbl>,
## # EducationField <fct>, EmployeeNumber <dbl>, EnvironmentSatisfaction <dbl>,
## # Gender <fct>, HourlyRate <dbl>, JobInvolvement <dbl>, JobLevel <dbl>,
## # JobRole <fct>, JobSatisfaction <dbl>, MaritalStatus <fct>,
## # MonthlyIncome <dbl>, MonthlyRate <dbl>, NumCompaniesWorked <dbl>,
## # OverTime <fct>, PercentSalaryHike <dbl>, PerformanceRating <dbl>, …performance_h2o <- h2o.performance(best_model, newdata = test_h2o)
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] "DeepLearning_grid_1_AutoML_7_20240423_184856_model_1"
##
## $model$type
## [1] "Key<Model>"
##
## $model$URL
## [1] "/3/Models/DeepLearning_grid_1_AutoML_7_20240423_184856_model_1"
##
##
## $model_checksum
## [1] "-2112073008793364328"
##
## $frame
## $frame$name
## [1] "test_tbl_sid_b48e_3"
##
##
## $frame_checksum
## [1] "-54413681510283746"
##
## $description
## NULL
##
## $scoring_time
## [1] 1.713913e+12
##
## $predictions
## NULL
##
## $MSE
## [1] 0.09887263
##
## $RMSE
## [1] 0.3144402
##
## $nobs
## [1] 369
##
## $custom_metric_name
## NULL
##
## $custom_metric_value
## [1] 0
##
## $r2
## [1] 0.2738621
##
## $logloss
## [1] 0.3784866
##
## $AUC
## [1] 0.8673139
##
## $pr_auc
## [1] 0.6842232
##
## $Gini
## [1] 0.7346278
##
## $mean_per_class_error
## [1] 0.2307443
##
## $domain
## [1] "No" "Yes"
##
## $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
## No Yes Error Rate
## No 290 19 0.0615 = 19 / 309
## Yes 24 36 0.4000 = 24 / 60
## Totals 314 55 0.1165 = 43 / 369
##
##
## $thresholds_and_metric_scores
## Metrics for Thresholds: Binomial metrics as a function of classification thresholds
## threshold f1 f2 f0point5 accuracy precision recall specificity
## 1 0.978409 0.032787 0.020747 0.078125 0.840108 1.000000 0.016667 1.000000
## 2 0.962851 0.064516 0.041322 0.147059 0.842818 1.000000 0.033333 1.000000
## 3 0.931140 0.095238 0.061728 0.208333 0.845528 1.000000 0.050000 1.000000
## 4 0.851139 0.125000 0.081967 0.263158 0.848238 1.000000 0.066667 1.000000
## 5 0.850000 0.153846 0.102041 0.312500 0.850949 1.000000 0.083333 1.000000
## absolute_mcc min_per_class_accuracy mean_per_class_accuracy tns fns fps tps
## 1 0.118299 0.016667 0.508333 309 59 0 1
## 2 0.167527 0.033333 0.516667 309 58 0 2
## 3 0.205458 0.050000 0.525000 309 57 0 3
## 4 0.237568 0.066667 0.533333 309 56 0 4
## 5 0.265973 0.083333 0.541667 309 55 0 5
## tnr fnr fpr tpr idx
## 1 1.000000 0.983333 0.000000 0.016667 0
## 2 1.000000 0.966667 0.000000 0.033333 1
## 3 1.000000 0.950000 0.000000 0.050000 2
## 4 1.000000 0.933333 0.000000 0.066667 3
## 5 1.000000 0.916667 0.000000 0.083333 4
##
## ---
## threshold f1 f2 f0point5 accuracy precision recall
## 364 0.000020 0.283019 0.496689 0.197889 0.176152 0.164835 1.000000
## 365 0.000014 0.282353 0.495868 0.197368 0.173442 0.164384 1.000000
## 366 0.000012 0.281690 0.495050 0.196850 0.170732 0.163934 1.000000
## 367 0.000007 0.281030 0.494234 0.196335 0.168022 0.163488 1.000000
## 368 0.000003 0.280374 0.493421 0.195822 0.165312 0.163043 1.000000
## 369 0.000001 0.279720 0.492611 0.195313 0.162602 0.162602 1.000000
## specificity absolute_mcc min_per_class_accuracy mean_per_class_accuracy tns
## 364 0.016181 0.051645 0.016181 0.508091 5
## 365 0.012945 0.046130 0.012945 0.506472 4
## 366 0.009709 0.039895 0.009709 0.504854 3
## 367 0.006472 0.032530 0.006472 0.503236 2
## 368 0.003236 0.022971 0.003236 0.501618 1
## 369 0.000000 0.000000 0.000000 0.500000 0
## fns fps tps tnr fnr fpr tpr idx
## 364 0 304 60 0.016181 0.000000 0.983819 1.000000 363
## 365 0 305 60 0.012945 0.000000 0.987055 1.000000 364
## 366 0 306 60 0.009709 0.000000 0.990291 1.000000 365
## 367 0 307 60 0.006472 0.000000 0.993528 1.000000 366
## 368 0 308 60 0.003236 0.000000 0.996764 1.000000 367
## 369 0 309 60 0.000000 0.000000 1.000000 1.000000 368
##
## $max_criteria_and_metric_scores
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.124479 0.626087 54
## 2 max f2 0.014039 0.669975 162
## 3 max f0point5 0.344917 0.718750 24
## 4 max accuracy 0.344917 0.894309 24
## 5 max precision 0.978409 1.000000 0
## 6 max recall 0.000335 1.000000 328
## 7 max specificity 0.978409 1.000000 0
## 8 max absolute_mcc 0.228765 0.565158 38
## 9 max min_per_class_accuracy 0.025684 0.766667 117
## 10 max mean_per_class_accuracy 0.053030 0.786893 80
## 11 max tns 0.978409 309.000000 0
## 12 max fns 0.978409 59.000000 0
## 13 max fps 0.000001 309.000000 368
## 14 max tps 0.000335 60.000000 328
## 15 max tnr 0.978409 1.000000 0
## 16 max fnr 0.978409 0.983333 0
## 17 max fpr 0.000001 1.000000 368
## 18 max tpr 0.000335 1.000000 328
##
## $gains_lift_table
## Gains/Lift Table: Avg response rate: 16.26 %, avg score: 7.32 %
## group cumulative_data_fraction lower_threshold lift cumulative_lift
## 1 1 0.01084011 0.850364 6.150000 6.150000
## 2 2 0.02168022 0.767188 6.150000 6.150000
## 3 3 0.03252033 0.652627 4.612500 5.637500
## 4 4 0.04065041 0.488102 6.150000 5.740000
## 5 5 0.05149051 0.444812 6.150000 5.826316
## 6 6 0.10027100 0.242952 3.416667 4.654054
## 7 7 0.15176152 0.123230 2.589474 3.953571
## 8 8 0.20054201 0.069858 1.366667 3.324324
## 9 9 0.30081301 0.029268 0.831081 2.493243
## 10 10 0.40108401 0.017165 0.997297 2.119257
## 11 11 0.50135501 0.009675 0.498649 1.795135
## 12 12 0.59891599 0.005170 0.512500 1.586199
## 13 13 0.69918699 0.002752 0.166216 1.382558
## 14 14 0.79945799 0.001224 0.000000 1.209153
## 15 15 0.89972900 0.000300 0.332432 1.111446
## 16 16 1.00000000 0.000001 0.000000 1.000000
## response_rate score cumulative_response_rate cumulative_score
## 1 1.000000 0.930885 1.000000 0.930885
## 2 1.000000 0.808199 1.000000 0.869542
## 3 0.750000 0.692175 0.916667 0.810420
## 4 1.000000 0.540434 0.933333 0.756422
## 5 1.000000 0.463258 0.947368 0.694704
## 6 0.555556 0.311268 0.756757 0.508167
## 7 0.421053 0.159296 0.642857 0.389800
## 8 0.222222 0.097595 0.540541 0.318723
## 9 0.135135 0.044030 0.405405 0.227159
## 10 0.162162 0.022159 0.344595 0.175909
## 11 0.081081 0.013239 0.291892 0.143375
## 12 0.083333 0.006818 0.257919 0.121130
## 13 0.027027 0.003879 0.224806 0.104315
## 14 0.000000 0.001812 0.196610 0.091459
## 15 0.054054 0.000683 0.180723 0.081342
## 16 0.000000 0.000132 0.162602 0.073199
## capture_rate cumulative_capture_rate gain cumulative_gain
## 1 0.066667 0.066667 515.000000 515.000000
## 2 0.066667 0.133333 515.000000 515.000000
## 3 0.050000 0.183333 361.250000 463.750000
## 4 0.050000 0.233333 515.000000 474.000000
## 5 0.066667 0.300000 515.000000 482.631579
## 6 0.166667 0.466667 241.666667 365.405405
## 7 0.133333 0.600000 158.947368 295.357143
## 8 0.066667 0.666667 36.666667 232.432432
## 9 0.083333 0.750000 -16.891892 149.324324
## 10 0.100000 0.850000 -0.270270 111.925676
## 11 0.050000 0.900000 -50.135135 79.513514
## 12 0.050000 0.950000 -48.750000 58.619910
## 13 0.016667 0.966667 -83.378378 38.255814
## 14 0.000000 0.966667 -100.000000 20.915254
## 15 0.033333 1.000000 -66.756757 11.144578
## 16 0.000000 1.000000 -100.000000 0.000000
## kolmogorov_smirnov
## 1 0.066667
## 2 0.133333
## 3 0.180097
## 4 0.230097
## 5 0.296764
## 6 0.437540
## 7 0.535275
## 8 0.556634
## 9 0.536408
## 10 0.536084
## 11 0.476052
## 12 0.419256
## 13 0.319417
## 14 0.199676
## 15 0.119741
## 16 0.000000h2o.auc(performance_h2o)## [1] 0.8673139h2o.confusionMatrix(performance_h2o)## Confusion Matrix (vertical: actual; across: predicted) for max f1 @ threshold = 0.12447934114241:
## No Yes Error Rate
## No 290 19 0.061489 =19/309
## Yes 24 36 0.400000 =24/60
## Totals 314 55 0.116531 =43/369h2o.metric(performance_h2o) %>% as_tibble() %>% filter(threshold %>% between (0.41, 0.42))## # A tibble: 2 × 20
## threshold f1 f2 f0point5 accuracy precision recall specificity
## <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl> <dbl>
## 1 0.420 0.512 0.401 0.709 0.892 0.955 0.35 0.997
## 2 0.419 0.506 0.399 0.691 0.889 0.913 0.35 0.994
## # ℹ 12 more variables: absolute_mcc <dbl>, min_per_class_accuracy <dbl>,
## # mean_per_class_accuracy <dbl>, tns <dbl>, fns <dbl>, fps <dbl>, tps <dbl>,
## # tnr <dbl>, fnr <dbl>, fpr <dbl>, tpr <dbl>, idx <int>