AutoBin

Skip to end of metadata
Go to start of metadata
Author Lily Dong
Creation date 06/01/2011
Firing policy all
Package org.seasr.meandre.components.transform.binning

DESCRIPTION

Overview: Automatically discretize scalar data for the Naive Bayesian classification model.Detailed Description: Given a table of Examples, define the bins for each scalar input column. When binning Uniformly, the number of bins is determined by Number of Bins property, and the boundaries of the bins are set so that they divide evenly over the range of the binned column.When binning by weight, Number of Items per Bin sets the size of each bin. The values are then binned so that in each bin there is the same number of items. For more details see description of property Number of Items per Bin'.Data Handling: This module does not modify the input data.Scalability: The module requires enough memory to make copies of each of the scalar input columns.

INPUTS

Name Description Example
exampleTable
Read a table of examples. It is type of org.seasr.datatypes.datamining.table.ExampleTable
 

OUTPUTS

Name Description Example
binTransform
Output ncsa.d2k.modules.core.datatype.table.transformations.BinTransform that contains all the information needed to discretize the Example Table
 
error
This port is used to output any unhandled errors encountered during the execution of this component
 

PROPERTIES

Name Description Default value
_debug_level
Controls the verbosity of debug messages printed by the component during execution.
Possible values are: off, severe, warning, info, config, fine, finer, finest, all
Append ',mirror' to any of the values above to mirror that output to the server logs.
info
nrOfBins
This property is used to set the number of bins absolutely. This will give equally spaced bins between the minimum and maximum for each scalar column. It must be higher than 1.
2
weight
This property is used to set the number of items per bin When binning by weight, this is the number of items that will go in each bin. However, the bins may contain more or fewer values than weight values, depending on how many items equal the bin limits. Typically the last bin will contain less or equal to weight values and the rest of the bins will contain a number that is equal or greater to weight values.It must be a positive integer.
1
method
This property is used to set the method for discretization. Select 1 to create binsby weight. This will create bins with an equal number of items in each slot. Select 0 to do uniform discretization by specifying the number of bins. This will result in equally spaced bins between the minimum and maximum for each scalar column. It must be 0 or 1.
0
_ignore_errors
Set to 'true' to ignore all unhandled exceptions and prevent the flow from being terminated. Setting this property to 'false' will result in the flow being terminated in the event an unhandled exception is thrown during the execution of this component
false
Labels:
binning binning Delete
transform transform Delete
Enter labels to add to this page:
Please wait 
Looking for a label? Just start typing.