RMOA (version 1.1.0)

trainMOA.MOA_recommender: Train a MOA recommender (e.g. a BRISMFPredictor) on a datastream

Description

Train a MOA recommender (e.g. a BRISMFPredictor) on a datastream

Usage

# S3 method for MOA_recommender
trainMOA(model, formula, data, subset,
  na.action = na.exclude, transFUN = identity, chunksize = 1000,
  trace = FALSE, options = list(maxruntime = +Inf), ...)

Value

An object of class MOA_trainedmodel which is a list with elements

• model: the updated supplied model object of class MOA_recommender

• call: the matched call

• na.action: the value of na.action

• terms: the terms in the model

• transFUN: the transFUN argument

Arguments

model

an object of class MOA_model, as returned by MOA_recommender, e.g. a BRISMFPredictor

formula

a symbolic description of the model to be fit. This should be of the form rating ~ userid + itemid, in that sequence. These should be columns in the data, where userid and itemid are integers and rating is numeric.

data

an object of class datastream set up e.g. with datastream_file, datastream_dataframe, datastream_matrix, datastream_ffdf or your own datastream.

subset

an optional vector specifying a subset of observations to be used in the fitting process.

na.action

a function which indicates what should happen when the data contain NAs. See model.frame for details. Defaults to na.exclude.

transFUN

a function which is used after obtaining chunksize number of rows from the data datastream before applying model.frame. Useful if you want to change the results get_points on the datastream (e.g. for making sure the factor levels are the same in each chunk of processing, some data cleaning, ...). Defaults to identity.

chunksize

the number of rows to obtain from the data datastream in one chunk of model processing. Defaults to 1000. Can be used to speed up things according to the backbone architecture of the datastream.

trace

logical, indicating to show information on how many datastream chunks are already processed as a message.

options

a names list of further options. Currently not used.

...

other arguments, currently not used yet

See Also

MOA_recommender, datastream_file, datastream_dataframe, datastream_matrix, datastream_ffdf, datastream, predict.MOA_trainedmodel

Examples

require(recommenderlab)
data(MovieLense)
x <- getData.frame(MovieLense)
x$itemid <- as.integer(as.factor(x$item))
x$userid <- as.integer(as.factor(x$user))
x$rating <- as.numeric(x$rating)
x <- head(x, 5000)

movielensestream <- datastream_dataframe(data=x)
movielensestream$get_points(3)

ctrl <- MOAoptions(model = "BRISMFPredictor", features = 10)
brism <- BRISMFPredictor(control=ctrl)
mymodel <- trainMOA(model = brism, rating ~ userid + itemid, 
 data = movielensestream, chunksize = 1000, trace=TRUE)
summary(mymodel$model)