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 zoomembers <- readr::read_csv('https://raw.githubusercontent.com/rfordatascience/tidytuesday/master/data/2020/2020-09-22/members.csv') %>%
mutate(across(where(is.character), factor))## Rows: 76519 Columns: 21
## ── Column specification ────────────────────────────────────────────────────────
## Delimiter: ","
## chr (10): expedition_id, member_id, peak_id, peak_name, season, sex, citizen...
## dbl (5): year, age, highpoint_metres, death_height_metres, injury_height_me...
## lgl (6): hired, success, solo, oxygen_used, died, injured
##
## 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(members, strata = "died")
train_tbl <- training(data_split)
test_tbl <- testing(data_split)recipe_obj <- recipe(died ~ ., data = train_tbl) %>%
# Remove zero variance variables
step_zv(all_predictors()) h2o.init()## Connection successful!
##
## R is connected to the H2O cluster:
## H2O cluster uptime: 30 minutes 24 seconds
## 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 1 day
## H2O cluster name: H2O_started_from_R_eliza_vpd819
## H2O cluster total nodes: 1
## H2O cluster total memory: 3.17 GB
## H2O cluster total cores: 16
## H2O cluster allowed cores: 16
## 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 1 day) 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 = 2345)##
|
| | 0%
|
|======================================================================| 100%train_h2o <- split.h2o[[1]]
valid_h2o <- split.h2o[[2]]
test_h2o <- as.h2o(test_tbl)##
|
| | 0%
|
|======================================================================| 100%y <- "died"
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,
exclude_algos = "DeepLearning",
nfolds = 5,
seed = 3456
)##
|
| | 0%
|
|=== | 4%
## 21:39:24.59: 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.
## 21:39:24.67: AutoML: XGBoost is not available; skipping it.
|
|========= | 13%
|
|=============== | 22%
|
|====================== | 31%
|
|============================ | 40%
|
|================================== | 49%
|
|========================================= | 58%
|
|=============================================== | 67%
|
|===================================================== | 76%
|
|=========================================================== | 85%
|
|================================================================== | 94%
|
|======================================================================| 100%models_h2o %>% typeof()## [1] "S4"models_h2o %>% slotNames()## [1] "project_name" "leader" "leaderboard" "event_log"
## [5] "modeling_steps" "training_info"models_h2o@leader## Model Details:
## ==============
##
## H2OBinomialModel: gbm
## Model ID: GBM_1_AutoML_4_20241121_213924
## Model Summary:
## number_of_trees number_of_internal_trees model_size_in_bytes min_depth
## 1 58 58 185796 1
## max_depth mean_depth min_leaves max_leaves mean_leaves
## 1 15 11.65517 2 208 88.72414
##
##
## H2OBinomialMetrics: gbm
## ** Reported on training data. **
##
## MSE: 1.591511e-06
## RMSE: 0.001261551
## LogLoss: 8.547813e-05
## Mean Per-Class Error: 0
## AUC: 1
## AUCPR: 1
## Gini: 1
## R^2: 0.9998863
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## FALSE TRUE Error Rate
## FALSE 48023 0 0.000000 =0/48023
## TRUE 0 692 0.000000 =0/692
## Totals 48023 692 0.000000 =0/48715
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.881735 1.000000 155
## 2 max f2 0.881735 1.000000 155
## 3 max f0point5 0.881735 1.000000 155
## 4 max accuracy 0.881735 1.000000 155
## 5 max precision 0.999993 1.000000 0
## 6 max recall 0.881735 1.000000 155
## 7 max specificity 0.999993 1.000000 0
## 8 max absolute_mcc 0.881735 1.000000 155
## 9 max min_per_class_accuracy 0.881735 1.000000 155
## 10 max mean_per_class_accuracy 0.881735 1.000000 155
## 11 max tns 0.999993 48023.000000 0
## 12 max fns 0.999993 689.000000 0
## 13 max fps 0.000049 48023.000000 399
## 14 max tps 0.881735 692.000000 155
## 15 max tnr 0.999993 1.000000 0
## 16 max fnr 0.999993 0.995665 0
## 17 max fpr 0.000049 1.000000 399
## 18 max tpr 0.881735 1.000000 155
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## H2OBinomialMetrics: gbm
## ** Reported on validation data. **
## ** Validation metrics **
##
## MSE: 5.933745e-06
## RMSE: 0.002435928
## LogLoss: 0.000103277
## Mean Per-Class Error: 0
## AUC: 1
## AUCPR: 1
## Gini: 1
## R^2: 0.9996365
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## FALSE TRUE Error Rate
## FALSE 8530 0 0.000000 =0/8530
## TRUE 0 144 0.000000 =0/144
## Totals 8530 144 0.000000 =0/8674
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.836076 1.000000 99
## 2 max f2 0.836076 1.000000 99
## 3 max f0point5 0.836076 1.000000 99
## 4 max accuracy 0.836076 1.000000 99
## 5 max precision 0.999996 1.000000 0
## 6 max recall 0.836076 1.000000 99
## 7 max specificity 0.999996 1.000000 0
## 8 max absolute_mcc 0.836076 1.000000 99
## 9 max min_per_class_accuracy 0.836076 1.000000 99
## 10 max mean_per_class_accuracy 0.836076 1.000000 99
## 11 max tns 0.999996 8530.000000 0
## 12 max fns 0.999996 143.000000 0
## 13 max fps 0.000048 8530.000000 399
## 14 max tps 0.836076 144.000000 99
## 15 max tnr 0.999996 1.000000 0
## 16 max fnr 0.999996 0.993056 0
## 17 max fpr 0.000048 1.000000 399
## 18 max tpr 0.836076 1.000000 99
##
## Gains/Lift Table: Extract with `h2o.gainsLift(<model>, <data>)` or `h2o.gainsLift(<model>, valid=<T/F>, xval=<T/F>)`
## H2OBinomialMetrics: gbm
## ** Reported on cross-validation data. **
## ** 5-fold cross-validation on training data (Metrics computed for combined holdout predictions) **
##
## MSE: 0.0001997517
## RMSE: 0.01413335
## LogLoss: 0.001075555
## Mean Per-Class Error: 0
## AUC: 1
## AUCPR: 1
## Gini: 1
## R^2: 0.9857354
##
## Confusion Matrix (vertical: actual; across: predicted) for F1-optimal threshold:
## FALSE TRUE Error Rate
## FALSE 48023 0 0.000000 =0/48023
## TRUE 0 692 0.000000 =0/692
## Totals 48023 692 0.000000 =0/48715
##
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 0.035777 1.000000 168
## 2 max f2 0.035777 1.000000 168
## 3 max f0point5 0.035777 1.000000 168
## 4 max accuracy 0.035777 1.000000 168
## 5 max precision 0.999999 1.000000 0
## 6 max recall 0.035777 1.000000 168
## 7 max specificity 0.999999 1.000000 0
## 8 max absolute_mcc 0.035777 1.000000 168
## 9 max min_per_class_accuracy 0.035777 1.000000 168
## 10 max mean_per_class_accuracy 0.035777 1.000000 168
## 11 max tns 0.999999 48023.000000 0
## 12 max fns 0.999999 642.000000 0
## 13 max fps 0.000001 48023.000000 399
## 14 max tps 0.035777 692.000000 168
## 15 max tnr 0.999999 1.000000 0
## 16 max fnr 0.999999 0.927746 0
## 17 max fpr 0.000001 1.000000 399
## 18 max tpr 0.035777 1.000000 168
##
## 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 1.000000 0.000000 1.000000 1.000000 1.000000
## auc 1.000000 0.000000 1.000000 1.000000 1.000000
## err 0.000000 0.000000 0.000000 0.000000 0.000000
## err_count 0.000000 0.000000 0.000000 0.000000 0.000000
## f0point5 1.000000 0.000000 1.000000 1.000000 1.000000
## f1 1.000000 0.000000 1.000000 1.000000 1.000000
## f2 1.000000 0.000000 1.000000 1.000000 1.000000
## lift_top_group 70.553345 3.681065 74.374050 69.099290 73.810610
## logloss 0.001076 0.000929 0.001440 0.002457 0.000119
## max_per_class_error 0.000000 0.000000 0.000000 0.000000 0.000000
## mcc 1.000000 0.000000 1.000000 1.000000 1.000000
## mean_per_class_accuracy 1.000000 0.000000 1.000000 1.000000 1.000000
## mean_per_class_error 0.000000 0.000000 0.000000 0.000000 0.000000
## mse 0.000200 0.000137 0.000165 0.000381 0.000028
## pr_auc 1.000000 0.000000 1.000000 1.000000 1.000000
## precision 1.000000 0.000000 1.000000 1.000000 1.000000
## r2 0.985981 0.009361 0.987583 0.973291 0.997941
## recall 1.000000 0.000000 1.000000 1.000000 1.000000
## rmse 0.013260 0.005469 0.012834 0.019518 0.005245
## specificity 1.000000 0.000000 1.000000 1.000000 1.000000
## cv_4_valid cv_5_valid
## accuracy 1.000000 1.000000
## auc 1.000000 1.000000
## err 0.000000 0.000000
## err_count 0.000000 0.000000
## f0point5 1.000000 1.000000
## f1 1.000000 1.000000
## f2 1.000000 1.000000
## lift_top_group 65.389260 70.093530
## logloss 0.000984 0.000378
## max_per_class_error 0.000000 0.000000
## mcc 1.000000 1.000000
## mean_per_class_accuracy 1.000000 1.000000
## mean_per_class_error 0.000000 0.000000
## mse 0.000288 0.000138
## pr_auc 1.000000 1.000000
## precision 1.000000 1.000000
## r2 0.980884 0.990206
## recall 1.000000 1.000000
## rmse 0.016967 0.011736
## specificity 1.000000 1.000000#h2o.getModel("StackedEnsemble_BestOfFamily_1_AutoML_3_20241121_213355") %>%
# h2o.saveModel("C:/Users/eliza/Desktop/PSU_DAT3100/11_module13/h2o_models/")
best_model <- h2o.loadModel("C:\\Users\\eliza\\Desktop\\PSU_DAT3100\\11_module13\\h2o_models\\StackedEnsemble_BestOfFamily_1_AutoML_3_20241121_213355")predictions <- h2o.predict(best_model, newdata = test_h2o)##
|
| | 0%
|
|======================================================================| 100%## Warning in doTryCatch(return(expr), name, parentenv, handler): Test/Validation
## dataset column 'expedition_id' has levels not trained on: ["AMAD00308",
## "AMAD02405", "AMAD07339", "AMAD11360", "AMAD11363", "AMAD12104", "AMAD12334",
## "AMAD13302", "AMAD17357", "AMAD98309", ...87 not listed..., "MANA16401",
## "MANA17102", "MANA17319", "MARD80301", "MERR11301", "MNSL14302", "PERI18401",
## "PUMO97303", "SNOW79301", "TUKU97103"]## Warning in doTryCatch(return(expr), name, parentenv, handler): Test/Validation
## dataset column 'member_id' has levels not trained on: ["ACHN18301-09",
## "AMAD00102-05", "AMAD00104-06", "AMAD00109-02", "AMAD00112-01", "AMAD00301-05",
## "AMAD00308-01", "AMAD00311-02", "AMAD00313-05", "AMAD00314-02", ...4781 not
## listed..., "YALU75101-07", "YALU81102-08", "YALU81102-17", "YALU81102-22",
## "YALU84302-05", "YALU85101-01", "YALU85101-16", "YALU89301-09", "YALU91301-06",
## "YAUP17101-04"]## Warning in doTryCatch(return(expr), name, parentenv, handler): Test/Validation
## dataset column 'peak_id' has levels not trained on: ["SNOW"]## Warning in doTryCatch(return(expr), name, parentenv, handler): Test/Validation
## dataset column 'peak_name' has levels not trained on: ["Snow Peak"]## Warning in doTryCatch(return(expr), name, parentenv, handler): Test/Validation
## dataset column 'citizenship' has levels not trained on: ["Canada/UK",
## "Chile/Sweden", "Netherlands/Switzerland", "Switzerland/Greece", "USA/Austria"]## Warning in doTryCatch(return(expr), name, parentenv, handler): Test/Validation
## dataset column 'expedition_role' has levels not trained on: ["BBC Producer",
## "BC Manager & Cook", "BC Manager (C1 only)", "Climber/Research", "Dep Leader
## (admin)", "Deputy Ldr/Exp Mgr", "Deputy Leader (Xixa)", "Deputy Leader III",
## "Everest-Lhotse Climber", "Exp Doctor (S side)", ...8 not listed..., "Leader of
## KPNU Grp", "Leader/Adv BC Manager", "Leader?", "Logistics Supervisor", "Naike",
## "Sandrine Marie Rose", "Ski Team Leader", "Student Leader", "Video
## Photographer", "Wireless Operator"]predictions_tbl <- predictions %>%
as_tibble()
predictions_tbl %>%
bind_cols(test_tbl)## # A tibble: 19,130 × 24
## predict FALSE. TRUE. expedition_id member_id peak_id peak_name year
## <fct> <dbl> <dbl> <fct> <fct> <fct> <fct> <dbl>
## 1 FALSE 1.00 2.75e-10 AMAD78301 AMAD78301-04 AMAD Ama Dablam 1978
## 2 FALSE 1.00 3.07e-10 AMAD79101 AMAD79101-05 AMAD Ama Dablam 1979
## 3 FALSE 1.00 3.81e-10 AMAD79101 AMAD79101-01 AMAD Ama Dablam 1979
## 4 FALSE 1.00 3.66e-10 AMAD79101 AMAD79101-06 AMAD Ama Dablam 1979
## 5 FALSE 1.00 3.77e-10 AMAD79101 AMAD79101-08 AMAD Ama Dablam 1979
## 6 FALSE 1.00 2.96e-10 AMAD79101 AMAD79101-02 AMAD Ama Dablam 1979
## 7 FALSE 1.00 2.99e-10 AMAD79101 AMAD79101-11 AMAD Ama Dablam 1979
## 8 FALSE 1.00 2.99e-10 AMAD79101 AMAD79101-13 AMAD Ama Dablam 1979
## 9 FALSE 1.00 3.77e-10 AMAD79101 AMAD79101-14 AMAD Ama Dablam 1979
## 10 FALSE 1.00 3.14e-10 AMAD79301 AMAD79301-12 AMAD Ama Dablam 1979
## # ℹ 19,120 more rows
## # ℹ 16 more variables: season <fct>, sex <fct>, age <dbl>, citizenship <fct>,
## # expedition_role <fct>, hired <lgl>, highpoint_metres <dbl>, success <lgl>,
## # solo <lgl>, oxygen_used <lgl>, died <lgl>, death_cause <fct>,
## # death_height_metres <dbl>, injured <lgl>, injury_type <fct>,
## # injury_height_metres <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] "StackedEnsemble_BestOfFamily_1_AutoML_3_20241121_213355"
##
## $model$type
## [1] "Key<Model>"
##
## $model$URL
## [1] "/3/Models/StackedEnsemble_BestOfFamily_1_AutoML_3_20241121_213355"
##
##
## $model_checksum
## [1] "-8897384442014169856"
##
## $frame
## $frame$name
## [1] "test_tbl_sid_804c_3"
##
##
## $frame_checksum
## [1] 9.199332e+14
##
## $description
## NULL
##
## $scoring_time
## [1] 1.732243e+12
##
## $predictions
## NULL
##
## $MSE
## [1] 1.62542e-11
##
## $RMSE
## [1] 4.03165e-06
##
## $nobs
## [1] 19130
##
## $custom_metric_name
## NULL
##
## $custom_metric_value
## [1] 0
##
## $r2
## [1] 1
##
## $logloss
## [1] 6.758657e-08
##
## $AUC
## [1] 1
##
## $pr_auc
## [1] 1
##
## $Gini
## [1] 1
##
## $mean_per_class_error
## [1] 0
##
## $domain
## [1] "FALSE" "TRUE"
##
## $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
## FALSE TRUE Error Rate
## FALSE 18860 0 0.0000 = 0 / 18,860
## TRUE 0 270 0.0000 = 0 / 270
## Totals 18860 270 0.0000 = 0 / 19,130
##
##
## $thresholds_and_metric_scores
## Metrics for Thresholds: Binomial metrics as a function of classification thresholds
## threshold f1 f2 f0point5 accuracy precision recall specificity
## 1 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
## 2 0.000538 0.998152 0.999260 0.997046 0.999948 0.996310 1.000000 0.999947
## 3 0.000071 0.996310 0.998521 0.994109 0.999895 0.992647 1.000000 0.999894
## 4 0.000052 0.994475 0.997783 0.991189 0.999843 0.989011 1.000000 0.999841
## 5 0.000052 0.992647 0.997046 0.988287 0.999791 0.985401 1.000000 0.999788
## absolute_mcc min_per_class_accuracy mean_per_class_accuracy tns fns fps tps
## 1 1.000000 1.000000 1.000000 18860 0 0 270
## 2 0.998127 0.999947 0.999973 18859 0 1 270
## 3 0.996264 0.999894 0.999947 18858 0 2 270
## 4 0.994411 0.999841 0.999920 18857 0 3 270
## 5 0.992569 0.999788 0.999894 18856 0 4 270
## tnr fnr fpr tpr idx
## 1 1.000000 0.000000 0.000000 1.000000 0
## 2 0.999947 0.000000 0.000053 1.000000 1
## 3 0.999894 0.000000 0.000106 1.000000 2
## 4 0.999841 0.000000 0.000159 1.000000 3
## 5 0.999788 0.000000 0.000212 1.000000 4
##
## ---
## threshold f1 f2 f0point5 accuracy precision recall
## 395 0.000000 0.031190 0.074491 0.019725 0.123210 0.015842 1.000000
## 396 0.000000 0.030193 0.072212 0.019087 0.093309 0.015328 1.000000
## 397 0.000000 0.029338 0.070254 0.018540 0.066074 0.014888 1.000000
## 398 0.000000 0.028653 0.068681 0.018103 0.043074 0.014535 1.000000
## 399 0.000000 0.028075 0.067352 0.017734 0.022791 0.014238 1.000000
## 400 0.000000 0.027835 0.066799 0.017580 0.014114 0.014114 1.000000
## specificity absolute_mcc min_per_class_accuracy mean_per_class_accuracy
## 395 0.110657 0.041870 0.110657 0.555329
## 396 0.080329 0.035089 0.080329 0.540164
## 397 0.052704 0.028011 0.052704 0.526352
## 398 0.029374 0.020663 0.029374 0.514687
## 399 0.008802 0.011194 0.008802 0.504401
## 400 0.000000 0.000000 0.000000 0.500000
## tns fns fps tps tnr fnr fpr tpr idx
## 395 2087 0 16773 270 0.110657 0.000000 0.889343 1.000000 394
## 396 1515 0 17345 270 0.080329 0.000000 0.919671 1.000000 395
## 397 994 0 17866 270 0.052704 0.000000 0.947296 1.000000 396
## 398 554 0 18306 270 0.029374 0.000000 0.970626 1.000000 397
## 399 166 0 18694 270 0.008802 0.000000 0.991198 1.000000 398
## 400 0 0 18860 270 0.000000 0.000000 1.000000 1.000000 399
##
## $max_criteria_and_metric_scores
## Maximum Metrics: Maximum metrics at their respective thresholds
## metric threshold value idx
## 1 max f1 1.000000 1.000000 0
## 2 max f2 1.000000 1.000000 0
## 3 max f0point5 1.000000 1.000000 0
## 4 max accuracy 1.000000 1.000000 0
## 5 max precision 1.000000 1.000000 0
## 6 max recall 1.000000 1.000000 0
## 7 max specificity 1.000000 1.000000 0
## 8 max absolute_mcc 1.000000 1.000000 0
## 9 max min_per_class_accuracy 1.000000 1.000000 0
## 10 max mean_per_class_accuracy 1.000000 1.000000 0
## 11 max tns 1.000000 18860.000000 0
## 12 max fns 1.000000 0.000000 0
## 13 max fps 0.000000 18860.000000 399
## 14 max tps 1.000000 270.000000 0
## 15 max tnr 1.000000 1.000000 0
## 16 max fnr 1.000000 0.000000 0
## 17 max fpr 0.000000 1.000000 399
## 18 max tpr 1.000000 1.000000 0
##
## $gains_lift_table
## Gains/Lift Table: Avg response rate: 1.41 %, avg score: 1.41 %
## group cumulative_data_fraction lower_threshold lift cumulative_lift
## 1 1 0.01359122 1.000000 70.851852 70.851852
## 2 2 0.02002091 0.000000 5.760313 49.947781
## 3 3 0.03000523 0.000000 0.000000 33.327526
## 4 4 0.04004182 0.000000 0.000000 24.973890
## 5 5 0.05002614 0.000000 0.000000 19.989551
## 6 6 0.10000000 0.000000 0.000000 10.000000
## 7 7 0.15002614 0.000000 0.000000 6.665505
## 8 8 0.20000000 0.000000 0.000000 5.000000
## 9 9 0.30000000 0.000000 0.000000 3.333333
## 10 10 0.40005227 0.000000 0.000000 2.499673
## 11 11 0.50000000 0.000000 0.000000 2.000000
## 12 12 0.60000000 0.000000 0.000000 1.666667
## 13 13 0.70000000 0.000000 0.000000 1.428571
## 14 14 0.80000000 0.000000 0.000000 1.250000
## 15 15 0.90000000 0.000000 0.000000 1.111111
## 16 16 1.00000000 0.000000 0.000000 1.000000
## response_rate score cumulative_response_rate cumulative_score
## 1 1.000000 1.000000 1.000000 1.000000
## 2 0.081301 0.081311 0.704961 0.704964
## 3 0.000000 0.000000 0.470383 0.470386
## 4 0.000000 0.000000 0.352480 0.352482
## 5 0.000000 0.000000 0.282132 0.282133
## 6 0.000000 0.000000 0.141140 0.141140
## 7 0.000000 0.000000 0.094077 0.094077
## 8 0.000000 0.000000 0.070570 0.070570
## 9 0.000000 0.000000 0.047047 0.047047
## 10 0.000000 0.000000 0.035280 0.035280
## 11 0.000000 0.000000 0.028228 0.028228
## 12 0.000000 0.000000 0.023523 0.023523
## 13 0.000000 0.000000 0.020163 0.020163
## 14 0.000000 0.000000 0.017642 0.017643
## 15 0.000000 0.000000 0.015682 0.015682
## 16 0.000000 0.000000 0.014114 0.014114
## capture_rate cumulative_capture_rate gain cumulative_gain
## 1 0.962963 0.962963 6985.185185 6985.185185
## 2 0.037037 1.000000 476.031316 4894.778068
## 3 0.000000 1.000000 -100.000000 3232.752613
## 4 0.000000 1.000000 -100.000000 2397.389034
## 5 0.000000 1.000000 -100.000000 1898.955068
## 6 0.000000 1.000000 -100.000000 900.000000
## 7 0.000000 1.000000 -100.000000 566.550523
## 8 0.000000 1.000000 -100.000000 400.000000
## 9 0.000000 1.000000 -100.000000 233.333333
## 10 0.000000 1.000000 -100.000000 149.967333
## 11 0.000000 1.000000 -100.000000 100.000000
## 12 0.000000 1.000000 -100.000000 66.666667
## 13 0.000000 1.000000 -100.000000 42.857143
## 14 0.000000 1.000000 -100.000000 25.000000
## 15 0.000000 1.000000 -100.000000 11.111111
## 16 0.000000 1.000000 -100.000000 0.000000
## kolmogorov_smirnov
## 1 0.962963
## 2 0.994008
## 3 0.983881
## 4 0.973701
## 5 0.963574
## 6 0.912884
## 7 0.862142
## 8 0.811453
## 9 0.710021
## 10 0.608537
## 11 0.507158
## 12 0.405726
## 13 0.304295
## 14 0.202863
## 15 0.101432
## 16 0.000000
##
## $residual_deviance
## [1] 0.002585862
##
## $null_deviance
## [1] 2837.674
##
## $AIC
## [1] 6.002586
##
## $loglikelihood
## [1] 0
##
## $null_degrees_of_freedom
## [1] 19129
##
## $residual_degrees_of_freedom
## [1] 19127h2o.auc(performance_h2o)## [1] 1h2o.confusionMatrix(performance_h2o)## Confusion Matrix (vertical: actual; across: predicted) for max f1 @ threshold = 0.99999999999871:
## FALSE TRUE Error Rate
## FALSE 18860 0 0.000000 =0/18860
## TRUE 0 270 0.000000 =0/270
## Totals 18860 270 0.000000 =0/19130h2o.metric(performance_h2o)## Metrics for Thresholds: Binomial metrics as a function of classification thresholds
## threshold f1 f2 f0point5 accuracy precision recall specificity
## 1 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000 1.000000
## 2 0.000538 0.998152 0.999260 0.997046 0.999948 0.996310 1.000000 0.999947
## 3 0.000071 0.996310 0.998521 0.994109 0.999895 0.992647 1.000000 0.999894
## 4 0.000052 0.994475 0.997783 0.991189 0.999843 0.989011 1.000000 0.999841
## 5 0.000052 0.992647 0.997046 0.988287 0.999791 0.985401 1.000000 0.999788
## absolute_mcc min_per_class_accuracy mean_per_class_accuracy tns fns fps tps
## 1 1.000000 1.000000 1.000000 18860 0 0 270
## 2 0.998127 0.999947 0.999973 18859 0 1 270
## 3 0.996264 0.999894 0.999947 18858 0 2 270
## 4 0.994411 0.999841 0.999920 18857 0 3 270
## 5 0.992569 0.999788 0.999894 18856 0 4 270
## tnr fnr fpr tpr idx
## 1 1.000000 0.000000 0.000000 1.000000 0
## 2 0.999947 0.000000 0.000053 1.000000 1
## 3 0.999894 0.000000 0.000106 1.000000 2
## 4 0.999841 0.000000 0.000159 1.000000 3
## 5 0.999788 0.000000 0.000212 1.000000 4
##
## ---
## threshold f1 f2 f0point5 accuracy precision recall
## 395 0.000000 0.031190 0.074491 0.019725 0.123210 0.015842 1.000000
## 396 0.000000 0.030193 0.072212 0.019087 0.093309 0.015328 1.000000
## 397 0.000000 0.029338 0.070254 0.018540 0.066074 0.014888 1.000000
## 398 0.000000 0.028653 0.068681 0.018103 0.043074 0.014535 1.000000
## 399 0.000000 0.028075 0.067352 0.017734 0.022791 0.014238 1.000000
## 400 0.000000 0.027835 0.066799 0.017580 0.014114 0.014114 1.000000
## specificity absolute_mcc min_per_class_accuracy mean_per_class_accuracy
## 395 0.110657 0.041870 0.110657 0.555329
## 396 0.080329 0.035089 0.080329 0.540164
## 397 0.052704 0.028011 0.052704 0.526352
## 398 0.029374 0.020663 0.029374 0.514687
## 399 0.008802 0.011194 0.008802 0.504401
## 400 0.000000 0.000000 0.000000 0.500000
## tns fns fps tps tnr fnr fpr tpr idx
## 395 2087 0 16773 270 0.110657 0.000000 0.889343 1.000000 394
## 396 1515 0 17345 270 0.080329 0.000000 0.919671 1.000000 395
## 397 994 0 17866 270 0.052704 0.000000 0.947296 1.000000 396
## 398 554 0 18306 270 0.029374 0.000000 0.970626 1.000000 397
## 399 166 0 18694 270 0.008802 0.000000 0.991198 1.000000 398
## 400 0 0 18860 270 0.000000 0.000000 1.000000 1.000000 399