Examples of weka.filters.unsupervised.attribute.AddExpression

Package weka.filters.unsupervised.attribute

Examples of weka.filters.unsupervised.attribute.AddExpression

weka.filters.unsupervised.attribute.AddExpression
An instance filter that creates a new attribute by applying a mathematical expression to existing attributes. The expression can contain attribute references and numeric constants. Supported operators are :
+, -, *, /, ^, log, abs, cos, exp, sqrt, floor, ceil, rint, tan, sin, (, )
Attributes are specified by prefixing with 'a', eg. a7 is attribute number 7 (starting from 1).
Example expression : a1^2*a5/log(a7*4.0).
Valid options are:
```
 -E <expression> Specify the expression to apply. Eg a1^2*a5/log(a7*4.0). Supported opperators: ,+, -, *, /, ^, log, abs, cos,  exp, sqrt, floor, ceil, rint, tan, sin, (, ) (default: a1^2)
```
```
 -N <name> Specify the name for the new attribute. (default is the expression provided with -E)
```
```
 -D Debug. Names attribute with the postfix parse of the expression.
```
@author Mark Hall (mhall@cs.waikato.ac.nz) @version $Revision: 5543 $

            }
          }


          avExpression += ")/" + denom;
          avAttName += ")";
          AddExpression addE = new AddExpression();
          addE.setName(avAttName);
          addE.setExpression(avExpression);
          addE.setInputFormat(insts);
          insts = Filter.useFilter(insts, addE);
          m_averagedLagMakers.add(addE);


          i += denom;
        } else {

View Full Code Here

       * = insts.instance(insts.numInstances() - 2); m_deltaTime =
       * first.value(timeStampIndex) - two.value(timeStampIndex);
       */


      m_timeIndexMakers = new ArrayList<Filter>();
      AddExpression addE = new AddExpression();
      addE.setName(timeStampName + "^2");
      addE.setExpression("a" + (timeStampIndex + 1) + "^2");
      addE.setInputFormat(insts);
      insts = Filter.useFilter(insts, addE);
      m_timeIndexMakers.add(addE);


      addE = new AddExpression();
      addE.setName(timeStampName + "^3");
      addE.setExpression("a" + (timeStampIndex + 1) + "^3");
      addE.setInputFormat(insts);
      insts = Filter.useFilter(insts, addE);
      m_timeIndexMakers.add(addE);
    }


    return insts;

View Full Code Here

      if (!(insts.attribute(i).name().startsWith("Lag_") || insts.attribute(i)
          .name().startsWith("Avg("))) {
        break;
      }


      AddExpression addE = new AddExpression();
      addE.setName(timeStampName + "*" + insts.attribute(i).name());
      addE.setExpression("a" + (timeStampIndex + 1) + "*a" + (i + 1));
      addE.setInputFormat(insts);
      insts = Filter.useFilter(insts, addE);
      m_timeLagCrossProductMakers.add(addE);
    }


    return insts;

View Full Code Here

    MultiFilter result = new MultiFilter();
    
    Filter[] filters = new Filter[2];
    filters[0] = new Add();
    ((Add) filters[0]).setAttributeIndex("last");
    filters[1] = new AddExpression();
    ((AddExpression) filters[1]).setExpression("a3+a6");
    
    result.setFilters(filters);
    
    return result;

View Full Code Here

    atts = new FastVector();
    atts.addElement(new Attribute("x"));
    
    m_RawData = new Instances(getRelationNameToUse(), atts, 0);


    m_Filter = new AddExpression();
    m_Filter.setName("y");
    m_Filter.setExpression(getExpression());
    m_Filter.setInputFormat(m_RawData);


    return super.defineDataFormat();

View Full Code Here

    atts = new FastVector();
    atts.addElement(new Attribute("x"));
    
    m_RawData = new Instances(getRelationNameToUse(), atts, 0);


    m_Filter = new AddExpression();
    m_Filter.setName("y");
    m_Filter.setExpression(getExpression());
    m_Filter.setInputFormat(m_RawData);


    return super.defineDataFormat();

View Full Code Here

TOP

Related Classes of weka.filters.unsupervised.attribute.AddExpression

weka.classifiers.timeseries.core.TSLagMaker

weka.core.Attribute

weka.core.AttributeExpression

weka.core.Capabilities

weka.core.DenseInstance

weka.core.Instance

weka.core.Instances

weka.core.Option

weka.core.SparseInstance

weka.datagenerators.classifiers.regression.Expression

All source code are property of their respective owners. Java is a trademark of Sun Microsystems, Inc and owned by ORACLE Inc. Contact coftware#gmail.com.