Skip to contents

Export h2o_automl's Results

Source: R/model_functions.R
export_results.Rd

Export RDS, TXT, POJO, MOJO and all results from h2o_automl().

Usage

export_results(
  results,
  thresh = 10,
  which = c("txt", "csv", "rds", "binary", "mojo", "plots", "dev", "production"),
  note = NA,
  subdir = NA,
  save = TRUE,
  seed = 0
)

Arguments

results

h2o_automl or h2o model

thresh

Integer. Threshold for selecting binary or regression models: this number is the threshold of unique values we should have in 'tag' (more than: regression; less than: classification)

which

Character vector. Select which file format to export: Possible values: txt, csv, rds, binary, mojo, plots. You might also use dev (txt, csv, rds) or production (binary, mojo) or simply don't use parameter to export everything

note

Character. Add a note to the txt file. Useful when lots of models are trained and saved to remember which one is which one

subdir

Character. In which directory do you wish to save the results?

save

Boolean. Do you wish to save/export results?

seed

Numeric. For reproducible results and random splits.

Value

No return value, called for side effects.

See also

Other Machine Learning: ROC(), conf_mat(), gain_lift(), h2o_automl(), h2o_predict_MOJO(), h2o_selectmodel(), impute(), iter_seeds(), lasso_vars(), model_metrics(), model_preprocess(), msplit()

Other Tools: autoline(), bind_files(), bring_api(), chr2num(), db_download(), db_upload(), export_plot(), files_functions(), font_exists(), formatColoured(), formatHTML(), get_credentials(), glued(), grepm(), h2o_selectmodel(), haveInternet(), image_metadata(), importxlsx(), ip_data(), json2vector(), list_cats(), listfiles(), mail_send(), markdown2df(), move_files(), msplit(), myip(), quiet(), read.file(), statusbar(), tic(), try_require(), updateLares(), warnifnot(), what_size()