RemixAutoML (version 0.2.8)

AutoXGBoostMultiClass: AutoXGBoostMultiClass is an automated XGBoost modeling framework with grid-tuning and model evaluation

Description

AutoXGBoostMultiClass is an automated XGBoost modeling framework with grid-tuning and model evaluation that runs a variety of steps. First, a stratified sampling (by the target variable) is done to create train and validation sets. Then, the function will run a random grid tune over N number of models and find which model is the best (a default model is always included in that set). Once the model is identified and built, several other outputs are generated: validation data with predictions, evaluation metrics, variable importance, and column names used in model fitting.

Usage

AutoXGBoostMultiClass(
  data,
  TrainOnFull = FALSE,
  ValidationData = NULL,
  TestData = NULL,
  TargetColumnName = NULL,
  FeatureColNames = NULL,
  IDcols = NULL,
  model_path = NULL,
  metadata_path = NULL,
  ModelID = "FirstModel",
  Objective = "multi:softmax",
  ReturnFactorLevels = TRUE,
  ReturnModelObjects = TRUE,
  SaveModelObjects = FALSE,
  Verbose = 0L,
  NumOfParDepPlots = 3L,
  NThreads = parallel::detectCores(),
  eval_metric = "merror",
  grid_eval_metric = "accuracy",
  TreeMethod = "hist",
  GridTune = FALSE,
  BaselineComparison = "default",
  MaxModelsInGrid = 10L,
  MaxRunsWithoutNewWinner = 20L,
  MaxRunMinutes = 24L * 60L,
  PassInGrid = NULL,
  Shuffles = 1L,
  Trees = 50L,
  eta = NULL,
  max_depth = NULL,
  min_child_weight = NULL,
  subsample = NULL,
  colsample_bytree = NULL
)

Arguments

data

This is your data set for training and testing your model

TrainOnFull

Set to TRUE to train on full data

ValidationData

This is your holdout data set used in modeling either refine your hyperparameters.

TestData

This is your holdout data set. Catboost using both training and validation data in the training process so you should evaluate out of sample performance with this data set.

TargetColumnName

Either supply the target column name OR the column number where the target is located (but not mixed types). Note that the target column needs to be a 0 | 1 numeric variable.

FeatureColNames

Either supply the feature column names OR the column number where the target is located (but not mixed types)

IDcols

A vector of column names or column numbers to keep in your data but not include in the modeling.

model_path

A character string of your path file to where you want your output saved

metadata_path

A character string of your path file to where you want your model evaluation output saved. If left NULL, all output will be saved to model_path.

ModelID

A character string to name your model and output

Objective

'multi:softmax'

ReturnFactorLevels

TRUE or FALSE. Set to FALSE to not return factor levels.

ReturnModelObjects

Set to TRUE to output all modeling objects (E.g. plots and evaluation metrics)

SaveModelObjects

Set to TRUE to return all modeling objects to your environment

Verbose

Set to 0 if you want to suppress model evaluation updates in training

NumOfParDepPlots

Tell the function the number of partial dependence calibration plots you want to create.

NThreads

Set the maximum number of threads you'd like to dedicate to the model run. E.g. 8

eval_metric

This is the metric used to identify best grid tuned model. Choose from "logloss","error","aucpr","auc"

grid_eval_metric

"accuracy", "logloss", "microauc"

TreeMethod

Choose from "hist", "gpu_hist"

GridTune

Set to TRUE to run a grid tuning procedure

BaselineComparison

Set to either "default" or "best". Default is to compare each successive model build to the baseline model using max trees (from function args). Best makes the comparison to the current best model.

MaxModelsInGrid

Number of models to test from grid options.

MaxRunsWithoutNewWinner

A number

MaxRunMinutes

In minutes

PassInGrid

Default is NULL. Provide a data.table of grid options from a previous run.

Shuffles

Numeric. List a number to let the program know how many times you want to shuffle the grids for grid tuning

Trees

Bandit grid partitioned. Supply a single value for non-grid tuning cases. Otherwise, supply a vector for the trees numbers you want to test. For running grid tuning, a NULL value supplied will mean these values are tested seq(1000L, 10000L, 1000L)

eta

Bandit grid partitioned. Supply a single value for non-grid tuning cases. Otherwise, supply a vector for the LearningRate values to test. For running grid tuning, a NULL value supplied will mean these values are tested c(0.01,0.02,0.03,0.04)

max_depth

Bandit grid partitioned. Number, or vector for depth to test. For running grid tuning, a NULL value supplied will mean these values are tested seq(4L, 16L, 2L)

min_child_weight

Number, or vector for min_child_weight to test. For running grid tuning, a NULL value supplied will mean these values are tested seq(1.0, 10.0, 1.0)

subsample

Number, or vector for subsample to test. For running grid tuning, a NULL value supplied will mean these values are tested seq(0.55, 1.0, 0.05)

colsample_bytree

Number, or vector for colsample_bytree to test. For running grid tuning, a NULL value supplied will mean these values are tested seq(0.55, 1.0, 0.05)

Value

Saves to file and returned in list: VariableImportance.csv, Model, ValidationData.csv, EvaluationMetrics.csv, GridCollect, GridList, and TargetLevels

See Also

Other Automated Supervised Learning - Multiclass Classification: AutoCatBoostMultiClass(), AutoH2oDRFMultiClass(), AutoH2oGAMMultiClass(), AutoH2oGBMMultiClass(), AutoH2oGLMMultiClass(), AutoH2oMLMultiClass()

Examples

# NOT RUN {
# Create some dummy correlated data
data <- RemixAutoML::FakeDataGenerator(
  Correlation = 0.85,
  N = 1000L,
  ID = 2L,
  ZIP = 0L,
  AddDate = FALSE,
  Classification = FALSE,
  MultiClass = TRUE)

# Run function
TestModel <- RemixAutoML::AutoXGBoostMultiClass(

    # GPU or CPU
    TreeMethod = "hist",
    NThreads = parallel::detectCores(),

    # Metadata arguments
    model_path = normalizePath("./"),
    metadata_path = file.path(normalizePath("./"),
      "R_Model_Testing"),
    ModelID = "Test_Model_1",
    ReturnFactorLevels = TRUE,
    ReturnModelObjects = TRUE,
    SaveModelObjects = FALSE,

    # Data arguments
    data = data,
    TrainOnFull = FALSE,
    ValidationData = NULL,
    TestData = NULL,
    TargetColumnName = "Adrian",
    FeatureColNames = names(data)[!names(data) %chin%
      c("IDcol_1", "IDcol_2","Adrian")],
    IDcols = c("IDcol_1","IDcol_2"),

    # Model evaluation
    eval_metric = "auc",
    Objective = 'multi:softmax',
    grid_eval_metric = "accuracy",
    NumOfParDepPlots = 3L,

    # Grid tuning arguments
    PassInGrid = NULL,
    GridTune = TRUE,
    BaselineComparison = "default",
    MaxModelsInGrid = 10L,
    MaxRunsWithoutNewWinner = 20L,
    MaxRunMinutes = 24L*60L,
    Verbose = 1L,

    # Trees, Depth, and LearningRate used in the bandit grid tuning
    # Must set Trees to a single value if you are not grid tuning
    # The ones below can be set to NULL
    #   and the values in the example will be used
    Shuffles = 1L,
    Trees = seq(50L, 500L, 50L),
    eta = seq(0.05,0.40,0.05),
    max_depth = seq(4L, 16L, 2L),
    min_child_weight = seq(1.0, 10.0, 1.0),
    subsample = seq(0.55, 1.0, 0.05),
    colsample_bytree = seq(0.55, 1.0, 0.05))
# }