BagModel

BagModel-class

show,BagModel-method

BagModel is an S4 class that contains the information of a bagging ensemble model.
Besides the base learning algorithms--<code>baseModels</code> --
BagModel class contains information regarding the
type of weak learners selected, the learned models and the aggregation method to apply in the 
test set for obtaining the final predictions.

Provides a set of functions that can be used to obtain better predictive performance on cost-sensitive and cost/benefits tasks (for both regression and classification). This includes re-sampling approaches that modify the original data set biasing it towards the user preferences.

Paula Branco

An Implementation of Re-Sampling Approaches to Utility-Based
Learning for Both Classification and Regression Tasks

Rita Ribeiro

Luis Torgo

BagModel-class function

Objects can be created by calls to the constructor
 <code>BagModel(...)</code>. The constructor requires a formula and a training set,
 the selected model type, the base models learned and the aggregation method
 to use for obtaining the final predictions.

Objects from BagModel Class

<dl></dl><dt><code>form</code></dt>
<dd>formula</dd>
<dt><code>train</code></dt>
<dd>training set used to train the <code>baseModels</code></dd>
<dt><code>learner</code></dt>
<dd>the weak learners type used</dd>
<dt><code>learner.pars</code></dt>
<dd>the parameters of the used learners</dd>
<dt><code>baseModels</code></dt>
<dd>a list of base learners</dd>
<dt><code>aggregation</code></dt>
<dd>the aggregation method
used to obtain the final predictions. For now, only the <code>average</code> method is available.</dd>
<dt><code>rel</code></dt>
<dd>optional information regarding the relevance function definde for the problem. Should be provided as a matrix.</dd>
<dt><code>thr.rel</code></dt>
<dd>Optional number setting the threshold on the relevance values.</dd><dt><code>quiet</code></dt>
<dd>logical specifying if development should be shown or not. Defaults to TRUE</dd>

Slots

Class "BagModel" — BagModel-class

<dl>

<dt><code>form</code></dt>
<dd>formula</dd>


<dt><code>train</code></dt>
<dd>training set used to train the <code>baseModels</code></dd>


<dt><code>learner</code></dt>
<dd>the weak learners type used</dd>


<dt><code>learner.pars</code></dt>
<dd>the parameters of the used learners</dd>


<dt><code>baseModels</code></dt>
<dd>a list of base learners</dd>


<dt><code>aggregation</code></dt>
<dd>the aggregation method
used to obtain the final predictions. For now, only the <code>average</code> method is available.</dd>


<dt><code>rel</code></dt>
<dd>optional information regarding the relevance function definde for the problem. Should be provided as a matrix.</dd>


<dt><code>thr.rel</code></dt>
<dd>Optional number setting the threshold on the relevance values.</dd>

<dt><code>quiet</code></dt>
<dd>logical specifying if development should be shown or not. Defaults to TRUE</dd>


</dl>

BagModel-class: Class "BagModel"

Description

Arguments

Objects from BagModel Class

Slots

See Also