FACT (version 0.1.1)

ClustPredictor: Clustering Predictor Object

Description

A ClustPredictor object holds any unsupervised clustering algorithm and the data to be used for analyzing the model. The interpretation methods in the FACT package need the clustering algorithm to be wrapped in a ClustPredictor object.

Super class

iml::Predictor -> ClustPredictor

Public fields

type

character(1)
Either partition for cluster assignments or prob for soft labels. Can be decided by chosen by the user when initializing the object. If NULL, it checks the the dimensions of y.

cnames

character
Is NULL, if hard labeling is used. If soft labels are used, column names of y are being transferred.

Methods

Public methods

• ClustPredictor$new()

• ClustPredictor$clone()

Inherited methods

• iml::Predictor$predict()
• iml::Predictor$print()

---

Method new()

Create a ClustPredictor object

Usage

ClustPredictor$new(
  model = NULL,
  data = NULL,
  predict.function = NULL,
  y = NULL,
  batch.size = 1000,
  type = NULL
)

Arguments

model

any
The trained clustering algorithm. Recommended are models from mlr3cluster. For other clustering algorithms predict functions need to be specified.

data

data.frame
The data to be used for analyzing the prediction model. Allowed column classes are: numeric, factor, integer, ordered and character

predict.function

function
The function to assign newdata. Only needed if model is not a model from mlr3cluster. The first argument of predict.fun has to be the model, the second the newdata:

function(model, newdata)

y

any
A integer vector representing the assigned clusters or a data.frame representing the soft labels per cluster assigned in columns.

batch.size

numeric(1)
The maximum number of rows to be input the model for prediction at once. Currently only respected for SMART.

type

character(1))
This argument is passed to the prediction function of the model. For soft label predictions, use type="prob". For hard label predictions, use type="partition". Consult the documentation or definition of the clustering algorithm you use to find which type options you have.

---

Method clone()

The objects of this class are cloneable with this method.

Usage

ClustPredictor$clone(deep = FALSE)

Arguments

deep

Whether to make a deep clone.

Details

A Cluster Predictor object is a container for the unsupervised prediction model and the data. This ensures that the clustering algorithm can be analyzed in a robust way. The Model inherits from iml::Predictor Object and adjusts this Object to contain unsupervised Methods.

Examples

require(factoextra)
require(FuzzyDBScan)
multishapes <- as.data.frame(multishapes[, 1:2])
eps = c(0, 0.2)
pts = c(3, 15)
res <- FuzzyDBScan$new(multishapes, eps, pts)
res$plot("x", "y")
# create hard label predictor
predict_part = function(model, newdata) model$predict(new_data = newdata, cmatrix = FALSE)$cluster
ClustPredictor$new(res, as.data.frame(multishapes), y = res$clusters,
                       predict.function = predict_part, type = "partition")
# create soft label predictor
predict_prob = function(model, newdata) model$predict(new_data = newdata)
ClustPredictor$new(res, as.data.frame(multishapes), y = res$results,
                               predict.function = predict_prob, type = "prob")