model_preprocess {lares}

Automate Data Preprocess for Modeling

Description

Pre-process your data before training a model. This is the prior step on the h2o_automl() function's pipeline. Enabling for other use cases when wanting too use any other framework, library, or custom algorithm.

Usage

model_preprocess(
  df,
  y = "tag",
  ignore = NULL,
  train_test = NA,
  split = 0.7,
  weight = NULL,
  target = "auto",
  balance = FALSE,
  impute = FALSE,
  no_outliers = TRUE,
  unique_train = TRUE,
  center = FALSE,
  scale = FALSE,
  thresh = 10,
  seed = 0,
  quiet = FALSE
)

Arguments

df

Dataframe. Dataframe containing all your data, including the dependent variable labeled as 'tag'. If you want to define which variable should be used instead, use the y parameter.

y

Character. Column name for dependent variable or response.

ignore

Character vector. Force columns for the model to ignore

train_test

Character. If needed, df's column name with 'test' and 'train' values to split

split

Numeric. Value between 0 and 1 to split as train/test datasets. Value is for training set. Set value to 1 to train with all available data and test with same data (cross-validation will still be used when training). If train_test is set, value will be overwritten with its real split rate.

weight

Column with observation weights. Giving some observation a weight of zero is equivalent to excluding it from the dataset; giving an observation a relative weight of 2 is equivalent to repeating that row twice. Negative weights are not allowed.

target

Value. Which is your target positive value? If set to 'auto', the target with largest mean(score) will be selected. Change the value to overwrite. Only used when binary categorical model.

balance

Boolean. Auto-balance train dataset with under-sampling?

impute

Boolean. Fill NA values with MICE?

no_outliers

Boolean/Numeric. Remove y's outliers from the dataset? Will remove those values that are farther than n standard deviations from the dependent variable's mean (Z-score). Set to TRUE for default (3) or numeric to set a different multiplier.

unique_train

Boolean. Keep only unique row observations for training data?

center, scale

Boolean. Using the base function scale, do you wish to center and/or scale all numerical values?

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)

seed

Integer. Set a seed for reproducibility. AutoML can only guarantee reproducibility if max_models is used because max_time is resource limited.

quiet

Boolean. Quiet all messages, warnings, recommendations?

Value

List. Contains original data.frame df, an index to identify which observations with be part of the train dataset train_index, and which model type should be model_type.

See Also

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

Examples

data(dft) # Titanic dataset

model_preprocess(dft, "Survived", balance = TRUE)

model_preprocess(dft, "Fare", split = 0.5, scale = TRUE)

model_preprocess(dft, "Pclass", ignore = c("Fare", "Cabin"))

model_preprocess(dft, "Pclass", quiet = TRUE)