ExhaustiveAttributeSelector (magpie API)

java.lang.Object
- magpie.attributes.selectors.BaseAttributeSelector
- - magpie.attributes.selectors.ExhaustiveAttributeSelector

All Implemented Interfaces:

java.io.Serializable, java.lang.Cloneable, Commandable, Options, Printable
```
public class ExhaustiveAttributeSelector
extends BaseAttributeSelector
```
From all possible subsets, find the one that leads to the highest CV score.
Iterates through all possible attribute subsets between two user-defined sizes. Each subset is rated by measuring which can be used to train a model with the highest score in cross-validation. For regression models, the subset with the lowest RMSE is selected and the model with the highest accuracy is used for classification models.

Usage: $<model> [-min_size <min>] [-max_size <max>] [-k_fold <k>] | [-random_cv <test_frac> <n_repeat>] | [-train]
model: BaseModel used to test attribute sets
min: Minimum attribute set size (default=1)
max: Maximum attribute set size (default=4)
k: Use k-fold CV to evaluate model: Number of folds to use during cross-validation
test_frac: Use random split CV for validation: Size of split for test set
n_repeat: Use random split CV for validation: Number of times to repeat test By default, class uses 10-fold CV. Can specify only one option of "-k_fold" (for k-fold CV), -random_cv (for multiple test/train set splits), and "-train" (for using training set score).

See Also:

Serialized Form

Nested Class Summary

Nested Classes
Modifier and Type Class and Description

static class ExhaustiveAttributeSelector.EvaluationMethod
List of methods used to evaluate model performance

Nested Classes
Modifier and Type	Class and Description
`static class`	`ExhaustiveAttributeSelector.EvaluationMethod` List of methods used to evaluate model performance

Field Summary

Fields
Modifier and Type	Field and Description
`protected int`	`KFolds` If test method is K-Fold: Number of folds used in cross-validation test
`protected int`	`MaxSubsetSize` Maximum attribute subset size
`protected int`	`MinSubsetSize` Minimum attribute subset size
`protected BaseModel`	`Model` Model used for cross-validation
`protected int`	`RandomTestCount` If method is random split, number of times to repeat test
`protected double`	`RandomTestFraction` If method is random split, fraction of entries withheld for test set
`protected java.util.Iterator<int[]>`	`SetIterator` Iterator over subsets of attributes to be tested
`protected ExhaustiveAttributeSelector.EvaluationMethod`	`TestMethod` Method used to evaluate models

Fields inherited from class magpie.attributes.selectors.BaseAttributeSelector
trained

Constructor Summary

Constructors
Constructor and Description

ExhaustiveAttributeSelector()

Constructors
Constructor and Description
`ExhaustiveAttributeSelector()`

Method Summary

All Methods Instance Methods Concrete Methods
Modifier and Type	Method and Description
`BaseAttributeSelector`	`clone()`
`java.lang.String`	`printDescription(boolean htmlFormat)` Print full name of object, and a simple description of the options.
`java.lang.String`	`printUsage()` Print out required format for options.
`protected void`	`setCombinationIterator(Dataset data)` Create an iterator over all attribute subsets being considered.
`void`	`setMaxSubsetSize(int size)` Set maximum size of subset to be tested
`void`	`setMinSubsetSize(int size)` Set minimum size of subset to be tested
`void`	`setModel(BaseModel model)` Set model used to test subset performance
`void`	`setNFolds(int k)` Set the number of folds used in K-fold cross-validation.
`void`	`setOptions(java.util.List<java.lang.Object> Options)` Set any options for this object.
`void`	`setRandomSplit(double test_split, int n_repeats)` Set the size of test set split and number of times to repeat CV.
`void`	`setTestMethod(ExhaustiveAttributeSelector.EvaluationMethod method)` Set method used to evaluate model performance
`protected java.util.List<java.lang.Integer>`	`train_protected(Dataset data)` Operation that actually does the work for training.

Methods inherited from class magpie.attributes.selectors.BaseAttributeSelector
about, applyAttributeSelection, getSelectionNames, getSelections, isTrained, printCommand, printSelections, run, runCommand, train

Methods inherited from class java.lang.Object
equals, finalize, getClass, hashCode, notify, notifyAll, toString, wait, wait, wait

- Field Detail
  - Model
```
protected BaseModel Model
```
    Model used for cross-validation
  - MinSubsetSize
```
protected int MinSubsetSize
```
    Minimum attribute subset size
  - MaxSubsetSize
```
protected int MaxSubsetSize
```
    Maximum attribute subset size
  - KFolds
```
protected int KFolds
```
    If test method is K-Fold: Number of folds used in cross-validation test
  - TestMethod
```
protected ExhaustiveAttributeSelector.EvaluationMethod TestMethod
```
    Method used to evaluate models
  - RandomTestFraction
```
protected double RandomTestFraction
```
    If method is random split, fraction of entries withheld for test set
  - RandomTestCount
```
protected int RandomTestCount
```
    If method is random split, number of times to repeat test
  - SetIterator
```
protected java.util.Iterator<int[]> SetIterator
```
    Iterator over subsets of attributes to be tested
- Constructor Detail
  - ExhaustiveAttributeSelector
```
public ExhaustiveAttributeSelector()
```
- Method Detail
  - clone
```
public BaseAttributeSelector clone()
```
    Overrides:
    
    clone in class BaseAttributeSelector
  - setOptions
```
public void setOptions(java.util.List<java.lang.Object> Options)
                throws java.lang.Exception
```
    Description copied from interface: Options
    
    Set any options for this object. If parsing the options fails, return an Exception with the usage as its message.
    
    Parameters:
    
    Options - Array of options as Objects - can be null
    
    Throws:
    
    java.lang.Exception - if problem with inputs
  - printUsage
```
public java.lang.String printUsage()
```
    Description copied from interface: Options
    
    Print out required format for options. For consistency, should start with "Usage: ".
    
    Returns:
    
    Usage for this function
  - setModel
```
public void setModel(BaseModel model)
```
    Set model used to test subset performance
    
    Parameters:
    
    model - Model template
  - setMinSubsetSize
```
public void setMinSubsetSize(int size)
```
    Set minimum size of subset to be tested
    
    Parameters:
    
    size - Desired minimum size
  - setMaxSubsetSize
```
public void setMaxSubsetSize(int size)
```
    Set maximum size of subset to be tested
    
    Parameters:
    
    size - Desired maximum size
  - setTestMethod
```
public void setTestMethod(ExhaustiveAttributeSelector.EvaluationMethod method)
```
    Set method used to evaluate model performance
    
    Parameters:
    
    method - Desired method
  - setNFolds
```
public void setNFolds(int k)
```
    Set the number of folds used in K-fold cross-validation. Will set K-fold CV to be the evaluation method
    
    Parameters:
    
    k - Number of folds
  - setRandomSplit
```
public void setRandomSplit(double test_split,
                           int n_repeats)
```
    Set the size of test set split and number of times to repeat CV. Will set Random Split CV to be the evaluation method
    
    Parameters:
    
    test_split -
    
    n_repeats -
  - train_protected
```
protected java.util.List<java.lang.Integer> train_protected(Dataset data)
```
    Description copied from class: BaseAttributeSelector
    
    Operation that actually does the work for training. Must generate a list of attribute IDs
    
    Specified by:
    
    train_protected in class BaseAttributeSelector
    
    Parameters:
    
    data - Dataset used to train selector
    
    Returns:
    
    List of indexes of selected attributes
  - setCombinationIterator
```
protected void setCombinationIterator(Dataset data)
```
    Create an iterator over all attribute subsets being considered. Set it as the current state
    
    Parameters:
    
    data - Dataset being used to train attribute selector
  - printDescription
```
public java.lang.String printDescription(boolean htmlFormat)
```
    Description copied from interface: Printable
    
    Print full name of object, and a simple description of the options. This description should print enough information to reproduce how this object was created, but not necessarily enough to run it (e.g., coefficients for the model).
    Example: For a model training a separate WekaRegression for intermetallics
    magpie.models.regression.SplitRegression
    Splitter: AllMetalsSplitter
    All Metals: magpie.models.regression.WekaRegression trees.REPTree
    Contains Nonmetal: magpie.regression.LASSORegreession -maxterms 2
    
    Specified by:
    
    printDescription in interface Printable
    
    Overrides:
    
    printDescription in class BaseAttributeSelector
    
    Parameters:
    
    htmlFormat - Whether format for output to an HTML page (e.g., <div> to create indentation) or for printing to screen.
    
    Returns:
    
    String describing the model
    
    See Also:
    
    #printModel()

Class ExhaustiveAttributeSelector

Nested Class Summary

Field Summary

Fields inherited from class magpie.attributes.selectors.BaseAttributeSelector

Constructor Summary

Method Summary

Methods inherited from class magpie.attributes.selectors.BaseAttributeSelector

Methods inherited from class java.lang.Object

Field Detail

Model

MinSubsetSize

MaxSubsetSize

KFolds

TestMethod

RandomTestFraction

RandomTestCount

SetIterator

Constructor Detail

ExhaustiveAttributeSelector

Method Detail

clone

setOptions

printUsage

setModel

setMinSubsetSize

setMaxSubsetSize

setTestMethod

setNFolds

setRandomSplit

train_protected

setCombinationIterator

printDescription