R/psvm.R

In h2o: R Interface for the 'H2O' Scalable Machine Learning Platform

# This file is auto-generated by h2o-3/h2o-bindings/bin/gen_R.py
# Copyright 2016 H2O.ai;  Apache License Version 2.0 (see LICENSE for details) 
#'
# -------------------------- Support Vector Machine -------------------------- #
#'
#' Trains a Support Vector Machine model on an H2O dataset
#' 
#' Alpha version. Supports only binomial classification problems. 
#'
#' @param x (Optional) A vector containing the names or indices of the predictor variables to use in building the model.
#'        If x is missing, then all columns except y are used.
#' @param y The name or column index of the response variable in the data. The response must be either a binary
#'        categorical/factor variable or a numeric variable with values -1/1 (for compatibility with SVMlight format).
#' @param training_frame Id of the training data frame.
#' @param model_id Destination id for this model; auto-generated if not specified.
#' @param validation_frame Id of the validation data frame.
#' @param ignore_const_cols \code{Logical}. Ignore constant columns. Defaults to TRUE.
#' @param hyper_param Penalty parameter C of the error term Defaults to 1.
#' @param kernel_type Type of used kernel Must be one of: "gaussian". Defaults to gaussian.
#' @param gamma Coefficient of the kernel (currently RBF gamma for gaussian kernel, -1 means 1/#features) Defaults to -1.
#' @param rank_ratio Desired rank of the ICF matrix expressed as an ration of number of input rows (-1 means use sqrt(#rows)).
#'        Defaults to -1.
#' @param positive_weight Weight of positive (+1) class of observations Defaults to 1.
#' @param negative_weight Weight of positive (-1) class of observations Defaults to 1.
#' @param disable_training_metrics \code{Logical}. Disable calculating training metrics (expensive on large datasets) Defaults to TRUE.
#' @param sv_threshold Threshold for accepting a candidate observation into the set of support vectors Defaults to 0.0001.
#' @param fact_threshold Convergence threshold of the Incomplete Cholesky Factorization (ICF) Defaults to 1e-05.
#' @param feasible_threshold Convergence threshold for primal-dual residuals in the IPM iteration Defaults to 0.001.
#' @param surrogate_gap_threshold Feasibility criterion of the surrogate duality gap (eta) Defaults to 0.001.
#' @param mu_factor Increasing factor mu Defaults to 10.
#' @param max_iterations Maximum number of iteration of the algorithm Defaults to 200.
#' @param seed Seed for random numbers (affects certain parts of the algo that are stochastic and those might or might not be enabled by default).
#'        Defaults to -1 (time-based random number).
#' @examples
#' \dontrun{
#' library(h2o)
#' h2o.init()
#' 
#' # Import the splice dataset
#' f <- "https://s3.amazonaws.com/h2o-public-test-data/smalldata/splice/splice.svm"
#' splice <- h2o.importFile(f)
#' 
#' # Train the Support Vector Machine model
#' svm_model <- h2o.psvm(gamma = 0.01, rank_ratio = 0.1,
#'                       y = "C1", training_frame = splice,
#'                       disable_training_metrics = FALSE)
#' }
#' @export
h2o.psvm <- function(x,
                     y,
                     training_frame,
                     model_id = NULL,
                     validation_frame = NULL,
                     ignore_const_cols = TRUE,
                     hyper_param = 1,
                     kernel_type = c("gaussian"),
                     gamma = -1,
                     rank_ratio = -1,
                     positive_weight = 1,
                     negative_weight = 1,
                     disable_training_metrics = TRUE,
                     sv_threshold = 0.0001,
                     fact_threshold = 1e-05,
                     feasible_threshold = 0.001,
                     surrogate_gap_threshold = 0.001,
                     mu_factor = 10,
                     max_iterations = 200,
                     seed = -1)
{
  # Validate required training_frame first and other frame args: should be a valid key or an H2OFrame object
  training_frame <- .validate.H2OFrame(training_frame, required=TRUE)
  validation_frame <- .validate.H2OFrame(validation_frame, required=FALSE)

  # Validate other required args
  # If x is missing, then assume user wants to use all columns as features.
  if (missing(x)) {
     if (is.numeric(y)) {
         x <- setdiff(col(training_frame), y)
     } else {
         x <- setdiff(colnames(training_frame), y)
     }
  }

  # Build parameter list to send to model builder
  parms <- list()
  parms$training_frame <- training_frame
  args <- .verify_dataxy(training_frame, x, y)
  parms$ignored_columns <- args$x_ignore
  parms$response_column <- args$y

  if (!missing(model_id))
    parms$model_id <- model_id
  if (!missing(validation_frame))
    parms$validation_frame <- validation_frame
  if (!missing(ignore_const_cols))
    parms$ignore_const_cols <- ignore_const_cols
  if (!missing(hyper_param))
    parms$hyper_param <- hyper_param
  if (!missing(kernel_type))
    parms$kernel_type <- kernel_type
  if (!missing(gamma))
    parms$gamma <- gamma
  if (!missing(rank_ratio))
    parms$rank_ratio <- rank_ratio
  if (!missing(positive_weight))
    parms$positive_weight <- positive_weight
  if (!missing(negative_weight))
    parms$negative_weight <- negative_weight
  if (!missing(disable_training_metrics))
    parms$disable_training_metrics <- disable_training_metrics
  if (!missing(sv_threshold))
    parms$sv_threshold <- sv_threshold
  if (!missing(fact_threshold))
    parms$fact_threshold <- fact_threshold
  if (!missing(feasible_threshold))
    parms$feasible_threshold <- feasible_threshold
  if (!missing(surrogate_gap_threshold))
    parms$surrogate_gap_threshold <- surrogate_gap_threshold
  if (!missing(mu_factor))
    parms$mu_factor <- mu_factor
  if (!missing(max_iterations))
    parms$max_iterations <- max_iterations
  if (!missing(seed))
    parms$seed <- seed

  # Error check and build model
  model <- .h2o.modelJob('psvm', parms, h2oRestApiVersion=3, verbose=FALSE)
  return(model)
}
.h2o.train_segments_psvm <- function(x,
                                     y,
                                     training_frame,
                                     validation_frame = NULL,
                                     ignore_const_cols = TRUE,
                                     hyper_param = 1,
                                     kernel_type = c("gaussian"),
                                     gamma = -1,
                                     rank_ratio = -1,
                                     positive_weight = 1,
                                     negative_weight = 1,
                                     disable_training_metrics = TRUE,
                                     sv_threshold = 0.0001,
                                     fact_threshold = 1e-05,
                                     feasible_threshold = 0.001,
                                     surrogate_gap_threshold = 0.001,
                                     mu_factor = 10,
                                     max_iterations = 200,
                                     seed = -1,
                                     segment_columns = NULL,
                                     segment_models_id = NULL,
                                     parallelism = 1)
{
  # formally define variables that were excluded from function parameters
  model_id <- NULL
  verbose <- NULL
  destination_key <- NULL
  # Validate required training_frame first and other frame args: should be a valid key or an H2OFrame object
  training_frame <- .validate.H2OFrame(training_frame, required=TRUE)
  validation_frame <- .validate.H2OFrame(validation_frame, required=FALSE)

  # Validate other required args
  # If x is missing, then assume user wants to use all columns as features.
  if (missing(x)) {
     if (is.numeric(y)) {
         x <- setdiff(col(training_frame), y)
     } else {
         x <- setdiff(colnames(training_frame), y)
     }
  }

  # Build parameter list to send to model builder
  parms <- list()
  parms$training_frame <- training_frame
  args <- .verify_dataxy(training_frame, x, y)
  parms$ignored_columns <- args$x_ignore
  parms$response_column <- args$y

  if (!missing(validation_frame))
    parms$validation_frame <- validation_frame
  if (!missing(ignore_const_cols))
    parms$ignore_const_cols <- ignore_const_cols
  if (!missing(hyper_param))
    parms$hyper_param <- hyper_param
  if (!missing(kernel_type))
    parms$kernel_type <- kernel_type
  if (!missing(gamma))
    parms$gamma <- gamma
  if (!missing(rank_ratio))
    parms$rank_ratio <- rank_ratio
  if (!missing(positive_weight))
    parms$positive_weight <- positive_weight
  if (!missing(negative_weight))
    parms$negative_weight <- negative_weight
  if (!missing(disable_training_metrics))
    parms$disable_training_metrics <- disable_training_metrics
  if (!missing(sv_threshold))
    parms$sv_threshold <- sv_threshold
  if (!missing(fact_threshold))
    parms$fact_threshold <- fact_threshold
  if (!missing(feasible_threshold))
    parms$feasible_threshold <- feasible_threshold
  if (!missing(surrogate_gap_threshold))
    parms$surrogate_gap_threshold <- surrogate_gap_threshold
  if (!missing(mu_factor))
    parms$mu_factor <- mu_factor
  if (!missing(max_iterations))
    parms$max_iterations <- max_iterations
  if (!missing(seed))
    parms$seed <- seed

  # Build segment-models specific parameters
  segment_parms <- list()
  if (!missing(segment_columns))
    segment_parms$segment_columns <- segment_columns
  if (!missing(segment_models_id))
    segment_parms$segment_models_id <- segment_models_id
  segment_parms$parallelism <- parallelism

  # Error check and build segment models
  segment_models <- .h2o.segmentModelsJob('psvm', segment_parms, parms, h2oRestApiVersion=3)
  return(segment_models)
}