spark.mlp {SparkR}R Documentation

Multilayer Perceptron Classification Model

Description

spark.mlp fits a multi-layer perceptron neural network model against a SparkDataFrame. Users can call summary to print a summary of the fitted model, predict to make predictions on new data, and write.ml/read.ml to save/load fitted models. Only categorical data is supported. For more details, see Multilayer Perceptron

Usage

spark.mlp(data, formula, ...)

## S4 method for signature 'SparkDataFrame,formula'
spark.mlp(data, formula, layers,
  blockSize = 128, solver = "l-bfgs", maxIter = 100, tol = 1e-06,
  stepSize = 0.03, seed = NULL, initialWeights = NULL)

## S4 method for signature 'MultilayerPerceptronClassificationModel'
predict(object, newData)

## S4 method for signature 'MultilayerPerceptronClassificationModel'
summary(object)

## S4 method for signature 'MultilayerPerceptronClassificationModel,character'
write.ml(object,
  path, overwrite = FALSE)

Arguments

data

a SparkDataFrame of observations and labels for model fitting.

formula

a symbolic description of the model to be fitted. Currently only a few formula operators are supported, including '~', '.', ':', '+', and '-'.

...

additional arguments passed to the method.

layers

integer vector containing the number of nodes for each layer.

blockSize

blockSize parameter.

solver

solver parameter, supported options: "gd" (minibatch gradient descent) or "l-bfgs".

maxIter

maximum iteration number.

tol

convergence tolerance of iterations.

stepSize

stepSize parameter.

seed

seed parameter for weights initialization.

initialWeights

initialWeights parameter for weights initialization, it should be a numeric vector.

object

a Multilayer Perceptron Classification Model fitted by spark.mlp

newData

a SparkDataFrame for testing.

path

the directory where the model is saved.

overwrite

overwrites or not if the output path already exists. Default is FALSE which means throw exception if the output path exists.

Value

spark.mlp returns a fitted Multilayer Perceptron Classification Model.

predict returns a SparkDataFrame containing predicted labeled in a column named "prediction".

summary returns summary information of the fitted model, which is a list. The list includes numOfInputs (number of inputs), numOfOutputs (number of outputs), layers (array of layer sizes including input and output layers), and weights (the weights of layers). For weights, it is a numeric vector with length equal to the expected given the architecture (i.e., for 8-10-2 network, 112 connection weights).

Note

spark.mlp since 2.1.0

predict(MultilayerPerceptronClassificationModel) since 2.1.0

summary(MultilayerPerceptronClassificationModel) since 2.1.0

write.ml(MultilayerPerceptronClassificationModel, character) since 2.1.0

See Also

read.ml

write.ml

Examples

## Not run: 
##D df <- read.df("data/mllib/sample_multiclass_classification_data.txt", source = "libsvm")
##D 
##D # fit a Multilayer Perceptron Classification Model
##D model <- spark.mlp(df, label ~ features, blockSize = 128, layers = c(4, 3), solver = "l-bfgs",
##D                    maxIter = 100, tol = 0.5, stepSize = 1, seed = 1,
##D                    initialWeights = c(0, 0, 0, 0, 0, 5, 5, 5, 5, 5, 9, 9, 9, 9, 9))
##D 
##D # get the summary of the model
##D summary(model)
##D 
##D # make predictions
##D predictions <- predict(model, df)
##D 
##D # save and load the model
##D path <- "path/to/model"
##D write.ml(model, path)
##D savedModel <- read.ml(path)
##D summary(savedModel)
## End(Not run)

[Package SparkR version 2.1.1 Index]