llama (version 0.9.2)

cluster: Cluster model

Description

Build a cluster model that predicts the algorithm to use based on the features of the problem.

Usage

cluster(clusterer = NULL, data = NULL,
    bestBy = "performance",
    pre = function(x, y=NULL) { list(features=x) },
    save.models = NA)

Arguments

clusterer

the mlr clustering function to use. See examples.

The argument can also be a list of such functions.

data

the data to use with training and test sets. The structure returned by one of the partitioning functions.

bestBy

the criteria by which to determine the best algorithm in a cluster. Can be one of "performance", "count", "successes". Optional. Defaults to "performance".

pre

a function to preprocess the data. Currently only normalize. Optional. Does nothing by default.

save.models

Whether to serialize and save the models trained during evaluation of the model. If not NA, will be used as a prefix for the file name.

Value

predictions

a data frame with the predictions for each instance and test set. The columns of the data frame are the instance ID columns (as determined by input), the algorithm, the score of the algorithm, and the iteration (e.g. the number of the fold for cross-validation). More than one prediction may be made for each instance and iteration. The score corresponds to the cumulative performance value for the algorithm of the cluster the instance was assigned to. That is, if bestBy is "performance", it is the sum of the performance over all training instances. If bestBy is "count", the score corresponds to the number of training instances that the respective algorithm was the best on, and if it is "successes" it corresponds to the number of training instances where the algorithm was successful. If more than one clustering algorithm is used, the score corresponds to the sum of all instances across all clusterers. If stacking is used, the prediction is simply the best algorithm with a score of 1.

predictor

a function that encapsulates the model learned on the entire data set. Can be called with data for the same features with the same feature names as the training data to obtain predictions in the same format as the predictions member.

models

the list of models trained on the entire data set. This is meant for debugging/inspection purposes and does not include any models used to combine predictions of individual models.

Details

cluster takes data and processes it using pre (if supplied). clusterer is called to cluster the data. For each cluster, the best algorithm is identified according to the criteria given in bestBy. If bestBy is "performance", the best algorithm is the one with the best overall performance across all instances in the cluster. If it is "count", the best algorithm is the one that has the best performance most often. If it is "successes", the best algorithm is the one with the highest number of successes across all instances in the cluster. The learned model is used to cluster the test data and predict algorithms accordingly.

The evaluation across the training and test sets will be parallelized automatically if a suitable backend for parallel computation is loaded. The parallelMap level is "llama.fold".

If a list of clusterers is supplied in clusterer, ensemble clustering is performed. That is, the models are trained and used to make predictions independently. For each instance, the final prediction is determined by majority vote of the predictions of the individual models -- the class that occurs most often is chosen. If the list given as clusterer contains a member .combine that is a function, it is assumed to be a classifier with the same properties as classifiers given to classify and will be used to combine the ensemble predictions instead of majority voting. This classifier is passed the original features and the predictions of the classifiers in the ensemble.

If all predictions of an underlying machine learning model are NA, the prediction will be NA for the algorithm and -Inf for the score if the performance value is to be maximised, Inf otherwise.

If save.models is not NA, the models trained during evaluation are serialized into files. Each file contains a list with members model (the mlr model), train.data (the mlr task with the training data), and test.data (the data frame with the test data used to make predictions). The file name starts with save.models, followed by the ID of the machine learning model, followed by "combined" if the model combines predictions of other models, followed by the number of the fold. Each model for each fold is saved in a different file.

See Also

classify, classifyPairs, regression, regressionPairs

Examples

# NOT RUN {
if(Sys.getenv("RUN_EXPENSIVE") == "true") {
data(satsolvers)
folds = cvFolds(satsolvers)

res = cluster(clusterer=makeLearner("cluster.XMeans"), data=folds, pre=normalize)
# the total number of successes
sum(successes(folds, res))
# predictions on the entire data set
res$predictor(satsolvers$data[satsolvers$features])

# determine best by number of successes
res = cluster(clusterer=makeLearner("cluster.XMeans"), data=folds,
    bestBy="successes", pre=normalize)
sum(successes(folds, res))

# ensemble clustering
rese = cluster(clusterer=list(makeLearner("cluster.XMeans"),
    makeLearner("cluster.SimpleKMeans"), makeLearner("cluster.EM")),
    data=folds, pre=normalize)

# ensemble clustering with a classifier to combine predictions
rese = cluster(clusterer=list(makeLearner("cluster.XMeans"),
    makeLearner("cluster.SimpleKMeans"), makeLearner("cluster.EM"),
    .combine=makeLearner("classif.J48")), data=folds, pre=normalize)
}
# }