Skip to main content

`exclude_algos`

  • Available in: AutoML
  • Hyperparameter: no

Description​

This option allows you to specify a list of algorithms that should not be included in an AutoML run during the model-building phase. This option defaults to None/Null, which means that all algorithms are included. However, if the include_algos option is used, then the AutoML run will include only those specified algorithms. Note that these two options cannot both be specified.

The algorithms that can be specified include:

  • DRF (including both the Random Forest and Extremely Randomized Trees (XRT) models)
  • GLM
  • XGBoost (XGBoost GBM)
  • GBM (H2O GBM)
  • DeepLearning (Fully connected multi-layer artificial neural network)
  • StackedEnsemble

Example​

library(h2o)
h2o.init()

# Import a sample binary outcome training set into H2O
train <- h2o.importFile("https://s3.amazonaws.com/h2o-public-test-data/smalldata/higgs/higgs_train_10k.csv")

# Identify predictors and response
y <- "response"
x <- setdiff(names(train), y)

# For binary classification, response should be a factor
train[, y] <- as.factor(train[, y])

# Train AutoML, omitting DeepLearning and DRF
aml <- h2o.automl(x = x, y = y,
                  training_frame = train,
                  max_runtime_secs = 30,
                  sort_metric = "logloss",
                  exclude_algos = c("DeepLearning", "DRF"))

# View the AutoML Leaderboard
lb <- aml@leaderboard
lb

                                             model_id       auc   logloss
1    StackedEnsemble_AllModels_AutoML_20190321_095825 0.7866967 0.5550255
2 StackedEnsemble_BestOfFamily_AutoML_20190321_095825 0.7848515 0.5569458
3                    XGBoost_1_AutoML_20190321_095825 0.7846668 0.5578654
4                    XGBoost_2_AutoML_20190321_095825 0.7820392 0.5586830
5           GLM_grid_1_AutoML_20190321_095825_model_1 0.6826481 0.6385205
  mean_per_class_error      rmse       mse
1            0.3309041 0.4338530 0.1882284
2            0.3231440 0.4346720 0.1889397
3            0.3324049 0.4349659 0.1891953
4            0.3269806 0.4356756 0.1898132
5            0.3972341 0.4726827 0.2234290

[5 rows x 6 columns]
Feedback