R Notebook: Quick-Start-R-H20

library(h2o)

h2o.init()

H2O is not running yet, starting it now...

Note:  In case of errors look at the following log files:
    /tmp/Rtmp101q6c/file18d1e8b7725/h2o_r3032219_started_from_r.out
    /tmp/Rtmp101q6c/file18d10572ba3/h2o_r3032219_started_from_r.err


Starting H2O JVM and connecting: ... Connection successful!

R is connected to the H2O cluster: 
    H2O cluster uptime:         2 seconds 708 milliseconds 
    H2O cluster timezone:       UTC 
    H2O data parsing timezone:  UTC 
    H2O cluster version:        3.44.0.3 
    H2O cluster version age:    2 years, 1 month and 16 days 
    H2O cluster name:           H2O_started_from_R_r3032219_fom207 
    H2O cluster total nodes:    1 
    H2O cluster total memory:   0.24 GB 
    H2O cluster total cores:    1 
    H2O cluster allowed cores:  1 
    H2O cluster healthy:        TRUE 
    H2O Connection ip:          localhost 
    H2O Connection port:        54321 
    H2O Connection proxy:       NA 
    H2O Internal Security:      FALSE 
    R Version:                  R version 4.5.2 (2025-10-31) 

h2o.init(nthreads = -1)

 Connection successful!

R is connected to the H2O cluster: 
    H2O cluster uptime:         1 minutes 55 seconds 
    H2O cluster timezone:       UTC 
    H2O data parsing timezone:  UTC 
    H2O cluster version:        3.44.0.3 
    H2O cluster version age:    2 years, 1 month and 16 days 
    H2O cluster name:           H2O_started_from_R_r3032219_fom207 
    H2O cluster total nodes:    1 
    H2O cluster total memory:   0.18 GB 
    H2O cluster total cores:    1 
    H2O cluster allowed cores:  1 
    H2O cluster healthy:        TRUE 
    H2O Connection ip:          localhost 
    H2O Connection port:        54321 
    H2O Connection proxy:       NA 
    H2O Internal Security:      FALSE 
    R Version:                  R version 4.5.2 (2025-10-31) 

datasets <- "https://raw.githubusercontent.com/DarrenCook/h2o/bk/datasets/"
data <- h2o.importFile(paste0(datasets, "iris_wheader.csv"))

  |                                                                                                       
  |                                                                                                 |   0%
  |                                                                                                       
  |=================================================================================================| 100%

y <- "class"
x <- setdiff(names(data), y)
parts <- h2o.splitFrame(data, 0.8)

#In R, h2o.splitFrame() takes an H2O frame and returns a list of the splits, which are assigned to train and #test, for readability:
train <- parts[[1]]
test <- parts[[2]]
m <- h2o.deeplearning(x, y, train)

  |                                                                                                       
  |                                                                                                 |   0%
  |                                                                                                       
  |==============================================================================                   |  80%
  |                                                                                                       
  |=================================================================================================| 100%

p <- h2o.predict(m, test)

  |                                                                                                       
  |                                                                                                 |   0%
  |                                                                                                       
  |=================================================================================================| 100%

h2o.mse(m)

[1] 0.1714094

h2o.confusionMatrix(m)

Confusion Matrix: Row labels: Actual class; Column labels: Predicted class

as.data.frame(p)

Finding Out Which Species H2O Model Got Wrong

as.data.frame( h2o.cbind(p$predict, test$class) )

The h2o model preformed very well overall, correctly identifying all Setosa and Virginica in the above table. The only errors occured within the bounderies of Virginica and Versicolor, where the model misclassified 4 Versicolor as Virginica.

What Percentage the H2O Model Got Right?

mean(p$predict == test$class)

[1] 0.8857143

The model guessed 88.6% of our unseen test samples correctly, and got 11.4% wrong.

Alternative to Find the Percentage…

h2o.performance(m, test)

H2OMultinomialMetrics: deeplearning

Test Set Metrics: 
=====================

MSE: (Extract with `h2o.mse`) 0.0968934
RMSE: (Extract with `h2o.rmse`) 0.311277
Logloss: (Extract with `h2o.logloss`) 0.3721349
Mean Per-Class Error: 0.1333333
AUC: (Extract with `h2o.auc`) NaN
AUCPR: (Extract with `h2o.aucpr`) NaN
Confusion Matrix: Extract with `h2o.confusionMatrix(<model>, <data>)`)
=========================================================================
Confusion Matrix: Row labels: Actual class; Column labels: Predicted class

Hit Ratio Table: Extract with `h2o.hit_ratio_table(<model>, <data>)`
=======================================================================
Top-3 Hit Ratios: 

NANANA