Menu
▾
▴
[Cdk-commits] SF.net SVN: cdk: [8280] trunk/cdk-taverna/src/main/java/org/openscience/cdk /applications/taverna/qsar/model/weka
|
From: <mar...@us...> - 2007-04-30 12:43:27
|
Revision: 8280
http://svn.sourceforge.net/cdk/?rev=8280&view=rev
Author: mario_baseda
Date: 2007-04-30 05:43:18 -0700 (Mon, 30 Apr 2007)
Log Message:
-----------
The classes "Selection" and "Standardise" now implements the LocalWorkerCDK interface
Modified Paths:
--------------
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateDensityBasedClustererModel.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateEMCluster.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateJ48WModel.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateKMeanCluster.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateLinearRegressionWModel.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateMultilayerPerceptronModel.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateNaiveBayesModel.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSMOModel.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSMOregModel.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSimpleLinearRegressionModel.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/Selection.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/Standardise.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterSupervisedResample.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterUnSupervisedRandomize.java
trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterUnSupervisedResample.java
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateDensityBasedClustererModel.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateDensityBasedClustererModel.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateDensityBasedClustererModel.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -26,6 +26,7 @@
package org.openscience.cdk.applications.taverna.qsar.model.weka;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -113,18 +114,33 @@
if (yMin == 0)yMin = 0.1;
if (yMax == 0)yMax = 0.9;
Standardise standardise = new Standardise(yMin, yMax);
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
for (int i = 0; i < (Integer)newInputs.get("moleculeCount"); i++){
y[i] = "unknown";
}
}
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
if (classAttrib == null) classAttrib = new String[]{"unknown"};
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
@@ -142,10 +158,26 @@
standardise = new Standardise(yMin, yMax);
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
- dbcm.setParameters(newMatix);
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
}
+ dbcm.setParameters(newMatrix);
+ }
} catch (Exception exception){
exception.getStackTrace();
}
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateEMCluster.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateEMCluster.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateEMCluster.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -27,6 +27,7 @@
package org.openscience.cdk.applications.taverna.qsar.model.weka;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -187,18 +188,33 @@
if (yMin == 0)yMin = 0.1;
if (yMax == 0)yMax = 0.9;
Standardise standardise = new Standardise(yMin, yMax);
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
for (int i = 0; i < (Integer)newInputs.get("moleculeCount"); i++){
y[i] = "unknown";
}
}
- if (classAttrib == null) classAttrib = new String[]{"unknown"};
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ if (classAttrib == null) classAttrib = new String[]{"unknown"};
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
@@ -219,10 +235,26 @@
standardise = new Standardise(yMin, yMax);
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
try{
- em.setParameters(newMatix);
+ em.setParameters(newMatrix);
em.logDensityPerClusterForInstance();
} catch (Exception exception){
exception.getStackTrace();
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateJ48WModel.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateJ48WModel.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateJ48WModel.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -27,6 +27,7 @@
package org.openscience.cdk.applications.taverna.qsar.model.weka;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -125,10 +126,13 @@
public Map execute(Map inputs) throws TaskExecutionException {
Selection selection = new Selection();
Standardise standardise = new Standardise();
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
for (int i = 0; i < (Integer)newInputs.get("moleculeCount"); i++){
@@ -136,7 +140,19 @@
}
}
if (classAttrib == null) classAttrib = new String[]{"unknown"};
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
@@ -155,9 +171,25 @@
standardise = new Standardise();
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
- j48.setParameters(newMatix);
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
+ j48.setParameters(newMatrix);
j48.predict();
}
} catch (Exception exception){
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateKMeanCluster.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateKMeanCluster.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateKMeanCluster.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -27,6 +27,7 @@
package org.openscience.cdk.applications.taverna.qsar.model.weka;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -118,10 +119,13 @@
if (yMin == 0)yMin = 0.1;
if (yMax == 0)yMax = 0.9;
Standardise standardise = new Standardise(yMin, yMax);
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
for (int i = 0; i < (Integer)newInputs.get("moleculeCount"); i++){
@@ -129,7 +133,19 @@
}
}
if (classAttrib == null) classAttrib = new String[]{"unknown"};
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
@@ -138,7 +154,6 @@
for (int q = 0; q < attrib.length; q++){
typAttrib[q] = Weka.NUMERIC;
}
-
try{
kMean.setData(attrib, typAttrib, classAttrib, y, matrix);
kMean.build();
@@ -150,10 +165,26 @@
standardise = new Standardise(yMin, yMax);
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
try{
- kMean.setParameters(newMatix);
+ kMean.setParameters(newMatrix);
kMean.clusterInstance();
} catch (Exception exception){
exception.getStackTrace();
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateLinearRegressionWModel.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateLinearRegressionWModel.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateLinearRegressionWModel.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -120,7 +120,7 @@
}
}
if (title == null || title.length == 0) {
- String[]titleTemp = selection.def(inputs, inputNames[0]);
+ String[]titleTemp = selection.def(inputs);
List titleList = new ArrayList();
for (int i = 0; i < titleTemp.length; i++){
if (titleTemp[i] != dependentAttrib[0]) {
@@ -137,18 +137,46 @@
}
}
if (cases == null) cases = standardise.def(title);
+ selection.setDescriptor(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (dependentAttrib == null){
dependentAttrib = new String[1];
dependentAttrib[0] = title[0];
}
- Map newYInputs = selection.selection(this.dependentAttrib, inputNames[0], inputs);
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
- Double[][] ymatrix = standardise.standardise(this.dependentAttrib, newYInputs, (Integer)((newYInputs.get("moleculeCount"))));
+ selection.setDescriptor(dependentAttrib);
+ Map newYInputs = selection.execute(inputs);
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
+ standardise.setDescriptor(dependentAttrib);
+ List ymatrixtmp = (List)((DataThing)(standardise.execute(newYInputs)).get("matrix")).getDataObject();
+ Double[][] ymatrix = new Double[ymatrixtmp.size()][((List)ymatrixtmp.get(0)).size()];
+ int z2 = 0 , s2 = 0;
+ for (Iterator iter = ymatrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ ymatrix[z2][s2] = val;
+ s2++;
+ }
+ z2++;
+ s2 = 0;
+ }
Double[] y = new Double[ymatrix.length];
- for (int z = 0; z < ymatrix.length; z++){
- y[z] = ymatrix[z][0];
+ for (int z3 = 0; z3 < ymatrix.length; z3++){
+ y[z3] = ymatrix[z3][0];
}
String[] attrib = new String[title.length+1];
for (int i = 0; i < title.length; i++) {
@@ -168,9 +196,25 @@
standardise = new Standardise();
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
- lrwm.setParameters(newMatix);
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
+ lrwm.setParameters(newMatrix);
lrwm.predict();
} catch (Exception exception){
exception.getStackTrace();
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateMultilayerPerceptronModel.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateMultilayerPerceptronModel.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateMultilayerPerceptronModel.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -27,6 +27,7 @@
import java.util.Arrays;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -132,18 +133,33 @@
if (yMin == 0)yMin = 0.1;
if (yMax == 0)yMax = 0.9;
Standardise standardise = new Standardise(yMin, yMax);
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
for (int i = 0; i < (Integer)newInputs.get("moleculeCount"); i++){
y[i] = "unknown";
}
}
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
if (classAttrib == null) classAttrib = new String[]{"unknown"};
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
@@ -164,10 +180,26 @@
standardise = new Standardise();
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
try{
- mpo.setParameters(newMatix);
+ mpo.setParameters(newMatrix);
mpo.predict();
outputs.put(outputNames()[0], new DataThing(getPredictPredicted()));
} catch (Exception exception){
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateNaiveBayesModel.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateNaiveBayesModel.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateNaiveBayesModel.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -27,6 +27,7 @@
import java.util.Arrays;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -116,10 +117,13 @@
if (yMin == 0)yMin = 0.1;
if (yMax == 0)yMax = 0.9;
Standardise standardise = new Standardise(yMin, yMax);
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
for (int i = 0; i < (Integer)newInputs.get("moleculeCount"); i++){
@@ -127,7 +131,19 @@
}
}
if (classAttrib == null) classAttrib = new String[]{"unknown"};
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
@@ -148,10 +164,26 @@
standardise = new Standardise();
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
try{
- nbm.setParameters(newMatix);
+ nbm.setParameters(newMatrix);
nbm.probabilities();
outputs.put(outputNames()[0], new DataThing(getProbabilities()));
} catch (Exception exception){
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSMOModel.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSMOModel.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSMOModel.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -28,6 +28,7 @@
import java.util.Arrays;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -132,10 +133,13 @@
if (yMin == 0)yMin = 0.1;
if (yMax == 0)yMax = 0.9;
Standardise standardise = new Standardise(yMin, yMax);
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
for (int i = 0; i < (Integer)newInputs.get("moleculeCount"); i++){
@@ -144,7 +148,19 @@
}
}
if (classAttrib == null) classAttrib = new String[]{"unknown_A", "unknown_B"};
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
@@ -162,13 +178,29 @@
}
if (inputs.containsKey(inputNames[1])){
selection = new Selection();
- standardise = new Standardise();
+ standardise = new Standardise(yMin, yMax);
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
try{
- smo.setParameters(newMatix);
+ smo.setParameters(newMatrix);
smo.probabilities();
outputs.put(outputNames()[0], new DataThing(getProbabilities()));
} catch (Exception exception){
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSMOregModel.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSMOregModel.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSMOregModel.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -123,7 +123,7 @@
}
}
if (title == null || title.length == 0) {
- String[]titleTemp = selection.def(inputs, inputNames[0]);
+ String[]titleTemp = selection.def(inputs);
List titleList = new ArrayList();
for (int i = 0; i < titleTemp.length; i++){
if (titleTemp[i] != dependentAttrib[0]) {
@@ -141,14 +141,41 @@
}
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
-
- Map newYInputs = selection.selection(this.dependentAttrib, inputNames[0], inputs);
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
- Double[][] ymatrix = standardise.standardise(this.dependentAttrib, newYInputs, (Integer)((newYInputs.get("moleculeCount"))));
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
+ selection.setDescriptor(dependentAttrib);
+ Map newYInputs = selection.execute(inputs);
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
+ standardise.setDescriptor(dependentAttrib);
+ List ymatrixtmp = (List)((DataThing)(standardise.execute(newYInputs)).get("matrix")).getDataObject();
+ Double[][] ymatrix = new Double[ymatrixtmp.size()][((List)ymatrixtmp.get(0)).size()];
+ int z2 = 0 , s2 = 0;
+ for (Iterator iter = ymatrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ ymatrix[z2][s2] = val;
+ s2++;
+ }
+ z2++;
+ s2 = 0;
+ }
Double[] y = new Double[ymatrix.length];
- for (int z = 0; z < ymatrix.length; z++){
- y[z] = ymatrix[z][0];
+ for (int z3 = 0; z3 < ymatrix.length; z3++){
+ y[z3] = ymatrix[z3][0];
}
String[] attrib = new String[title.length+1];
for (int i = 0; i < title.length; i++) {
@@ -167,9 +194,25 @@
standardise = new Standardise(yMin, yMax);
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
- smoregm.setParameters(newMatix);
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
+ smoregm.setParameters(newMatrix);
smoregm.classifyInstance();
outputs.put(outputNames()[0], new DataThing(getClassification()));
} }catch (Exception exception){
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSimpleLinearRegressionModel.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSimpleLinearRegressionModel.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/GenerateSimpleLinearRegressionModel.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -112,7 +112,7 @@
}
}
if (title == null || title.length == 0) {
- String[]titleTemp = selection.def(inputs, inputNames[0]);
+ String[]titleTemp = selection.def(inputs);
List titleList = new ArrayList();
for (int i = 0; i < titleTemp.length; i++){
if (titleTemp[i] != dependentAttrib[0]) {
@@ -130,13 +130,41 @@
}
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
- Map newYInputs = selection.selection(this.dependentAttrib, inputNames[0], inputs);
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
- Double[][] ymatrix = standardise.standardise(this.dependentAttrib, newYInputs, (Integer)((newYInputs.get("moleculeCount"))));
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
+ selection.setDescriptor(dependentAttrib);
+ Map newYInputs = selection.execute(inputs);
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
+ standardise.setDescriptor(dependentAttrib);
+ List ymatrixtmp = (List)((DataThing)(standardise.execute(newYInputs)).get("matrix")).getDataObject();
+ Double[][] ymatrix = new Double[ymatrixtmp.size()][((List)ymatrixtmp.get(0)).size()];
+ int z2 = 0 , s2 = 0;
+ for (Iterator iter = ymatrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ ymatrix[z2][s2] = val;
+ s2++;
+ }
+ z2++;
+ s2 = 0;
+ }
Double[] y = new Double[ymatrix.length];
- for (int z = 0; z < ymatrix.length; z++){
- y[z] = ymatrix[z][0];
+ for (int z3 = 0; z3 < ymatrix.length; z3++){
+ y[z3] = ymatrix[z3][0];
}
String[] attrib = new String[title.length+1];
for (int i = 0; i < title.length; i++) {
@@ -154,9 +182,25 @@
standardise = new Standardise();
standardise.setCases(cases);
newInputs.clear();
- newInputs = selection.selection(title, inputNames[1], inputs);
- Double[][] newMatix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
- slrwm.setParameters(newMatix);
+ selection.setDescriptor(title);
+ standardise.setDescriptor(title);
+ selection.setFiles(1);
+ newInputs = selection.execute(inputs);
+ List newMatrixl = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] newMatrix = new Double[newMatrixl.size()][((List)newMatrixl.get(0)).size()];
+ z = 0;
+ s = 0;
+ for (Iterator iter = newMatrixl.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ newMatrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
+ slrwm.setParameters(newMatrix);
slrwm.predict();
} catch (Exception exception){
exception.getStackTrace();
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/Selection.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/Selection.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/Selection.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -34,6 +34,7 @@
import org.embl.ebi.escience.baclava.DataThing;
import org.openscience.cdk.applications.taverna.CMLChemFile;
+import org.openscience.cdk.applications.taverna.LocalWorkerCDK;
import org.openscience.cdk.interfaces.IAtomContainer;
import org.openscience.cdk.qsar.DescriptorValue;
import org.openscience.cdk.qsar.result.DoubleArrayResult;
@@ -51,29 +52,50 @@
* @cdk.keyword clusterers
* @cdk.keyword SimpleKMeans
*/
-public class Selection {
+public class Selection implements LocalWorkerCDK{
+ // The attributs of which the molecule descriptors should be select
+ private String[] descriptor;
+ // 0 means NOT new files
+ // 1 means new files
+ private int newFiles = 0;
+ private String[] inputNames = new String[] { "inputChemFiles", "newInputChemFiles" };
+
+ private String[] outputNames = new String[] { "variousSpecification", "moleculeCount"};
+
+ public String[] inputNames() {
+ return inputNames;
+ }
+ public String[] inputTypes() {
+ return new String[] {LocalWorkerCDK.CMLChemFileList, LocalWorkerCDK.CMLChemFileList};
+ }
+
+ public String[] outputNames() {
+ return outputNames;
+ }
+
+ public String[] outputTypes() {
+ return new String[] {LocalWorkerCDK.DOUBLE, LocalWorkerCDK.INT, LocalWorkerCDK.INT};
+ }
/**
* Constructor of the selection object
*/
public Selection() {}
- /** This method select descriptors of a map of molecules about chose descriptors before
- * @param descriptor The attributs of which the molecule descriptors should be select
- * @param inputNames The map key which contains the molecules
+ /** This method select descriptors of a map of molecules about chose descriptors before
* @param inputs Map which contains the CML Chemfiles with the molecules
* @return subMap Map which contains the selected molecule descriptors with their values and the number of
* molecules in the map.
*/
- public Map selection(String[] descriptor, String inputNames, Map inputs) {
+ public Map execute(Map inputs) {
Map subMap = new HashMap();
int moleculeCount = 0;
for (String string : descriptor) {
subMap.put(string, new ArrayList());
}
List inputList = null;
- if (inputs.get(inputNames) != null) {
- inputList = (List) ((DataThing) (inputs.get(inputNames))).getDataObject();
+ if (inputs.get(this.inputNames[newFiles]) != null) {
+ inputList = (List) ((DataThing) (inputs.get(this.inputNames[newFiles]))).getDataObject();
} else {
return null;
}
@@ -114,7 +136,7 @@
}
}
}
- subMap.put("moleculeCount", moleculeCount);
+ subMap.put(outputNames[1], moleculeCount);
return subMap;
}
@@ -125,11 +147,11 @@
* @param inputNames The map key which contains the molecules
* @return descriptor A String[] which contains all Descriptor titles of the molecules in the CMLChemfiles
*/
- public String[] def(Map map, String inputNames){
+ public String[] def(Map map){
List descriptorList = new ArrayList();
int count = 0;
- if (map.get(inputNames) != null) {
- CMLChemFile file = (CMLChemFile)(((List) ((DataThing) (map.get(inputNames))).getDataObject()).get(0));
+ if (map.get(this.inputNames[0]) != null) {
+ CMLChemFile file = (CMLChemFile)(((List) ((DataThing) (map.get(inputNames[0]))).getDataObject()).get(0));
IAtomContainer molecules = (IAtomContainer)((ChemFileManipulator.getAllAtomContainers(file)).get(0));
for (Enumeration e = (molecules.getProperties()).elements(); e.hasMoreElements();) {
Object p = (Object) e.nextElement();
@@ -148,7 +170,6 @@
}
}
}
-
}else {
return null;
}
@@ -156,9 +177,23 @@
int i = 0;
for (Iterator iter = descriptorList.iterator(); iter.hasNext();) {
String element = (String) iter.next();
- descriptor[i] = element;
+ this.descriptor[i] = element;
i++;
}
- return descriptor;
+ return this.descriptor;
}
+
+ /**Set the attributs. The descriptor values will selected about this attributs
+ * @param title A string[] which contains the attributs
+ */
+ public void setDescriptor(String[] title){
+ this.descriptor = title;
+ }
+
+ /**Choose the input name
+ * @param tmp An int which chose the input name. 0 = inputChemFiles; 1 = newInputChemFiles
+ */
+ public void setFiles(int tmp){
+ newFiles = tmp;
+ }
}
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/Standardise.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/Standardise.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/Standardise.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -32,6 +32,9 @@
import java.util.HashMap;
import java.util.ArrayList;
+import org.embl.ebi.escience.baclava.DataThing;
+import org.openscience.cdk.applications.taverna.LocalWorkerCDK;
+
/** A class which standardise descriptor values about a minimum and maximum value.
* @author Mario Baseda
* @cdk.require cdk-svn.jar
@@ -42,11 +45,31 @@
* @cdk.keyword clusterers
* @cdk.keyword SimpleKMeans
*/
-public class Standardise {
+public class Standardise implements LocalWorkerCDK{
private double yMin, yMax;
private Map cases = null;
private boolean standard;
+ private String[] descriptor;
+
+private String[] inputNames = new String[] {"variousSpecification", "moleculeCount"};
+
+ private String[] outputNames = new String[] {"matrix"};
+
+ public String[] inputNames() {
+ return inputNames;
+ }
+ public String[] inputTypes() {
+ return new String[] {LocalWorkerCDK.DOUBLE, LocalWorkerCDK.INT, LocalWorkerCDK.INT};
+ }
+ public String[] outputNames() {
+ return outputNames;
+ }
+
+ public String[] outputTypes() {
+ return new String[] {LocalWorkerCDK.UNTYPED_ARRAY};
+ }
+
/** Constructur of the Standardise object
* @param yMin The minimum value for the standardization
* @param yMax The maximum value for the standardization
@@ -65,15 +88,19 @@
this.standard = false;
}
/** This method standardise descriptor values about a minimum and maximum value, which was set before.
- * @param map A Map which contains the different descriptors of the molecules
- * @param moleculeCount The number of molecules and so the number of values of each descriptor
+ * @param map A Map which contains the different descriptors of the molecules
* @return A matrix as 2D Double array, which contains the standardise descriptor values. Each line for each molecule
* and each column for each descriptor
*/
- public Double[][] standardise(String[] descriptor, Map map, int moleculeCount) {
+ public Map execute(Map map) {
+
+ int moleculeCount = (Integer)map.get("moleculeCount");
map.remove("moleculeCount");
- Double[][] matrix = new Double[moleculeCount][map.size()];
- int z = 0, s = 0;
+ List matrixl = new ArrayList();
+ for (int i = 0; i < moleculeCount; i++) {
+ matrixl.add(new ArrayList());
+ }
+ int z = 0;
for (Object object : descriptor) {
List tmpList = new ArrayList();
for (Iterator iter = ((List)map.get(object)).iterator(); iter.hasNext();) {
@@ -86,7 +113,7 @@
for (Iterator iter = ((List)(map.get((String)object))).iterator(); iter.hasNext();){
List temp = ((List)iter.next());
if ((temp.get(0) instanceof String) && ((String)temp.get(0) == "not available")) {
- matrix[z][s] = new Double (0.0d / 0.0);
+ ((List)matrixl.get(z)).add(new Double (0.0d / 0.0));
z++;
continue;
}
@@ -96,16 +123,17 @@
Arrays.sort(tmp);
double std = (re.doubleValue() * (yMax - yMin)+ ((Double)tmp[tmp.length-1]).doubleValue() * yMin - yMax * ((Double)tmp[0]).doubleValue()) /
(((Double)tmp[tmp.length-1]).doubleValue() - ((Double)tmp[0]).doubleValue());
- matrix[z][s] = new Double (std);
+ ((List)matrixl.get(z)).add(new Double (std));
z++;
- } else { matrix[z][s] = re;
+ } else { ((List)matrixl.get(z)).add(re);
z++;
}
}
- s++;
z = 0;
}
- return matrix;
+ Map subMap = new HashMap();
+ subMap.put(outputNames()[0],new DataThing(matrixl));
+ return subMap;
}
/**This method appeals a pre processing method to "calculate" a molecule discriptor value about the atom discriptor values
@@ -252,4 +280,11 @@
}
return map;
}
+
+ /**Set the attributs. The descriptor values will selected about this attributs
+ * @param title A string[] which contains the attributs
+ */
+ public void setDescriptor(String[] title){
+ descriptor = title;
+ }
}
\ No newline at end of file
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterSupervisedResample.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterSupervisedResample.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterSupervisedResample.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -26,6 +26,7 @@
package org.openscience.cdk.applications.taverna.qsar.model.weka;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -116,10 +117,13 @@
public Map execute(Map inputs) throws TaskExecutionException {
Selection selection = new Selection();
Standardise standardise = new Standardise();
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (classAttrib == null) classAttrib = new String[]{"unknown"};
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
@@ -127,7 +131,19 @@
y[i] = "unknown";
}
}
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterUnSupervisedRandomize.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterUnSupervisedRandomize.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterUnSupervisedRandomize.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -26,6 +26,7 @@
package org.openscience.cdk.applications.taverna.qsar.model.weka;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -107,10 +108,13 @@
public Map execute(Map inputs) throws TaskExecutionException {
Selection selection = new Selection();
Standardise standardise = new Standardise();
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
for (int i = 0; i < (Integer)newInputs.get("moleculeCount"); i++){
@@ -118,7 +122,19 @@
}
}
if (classAttrib == null) classAttrib = new String[]{"unknown"};
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
Modified: trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterUnSupervisedResample.java
===================================================================
--- trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterUnSupervisedResample.java 2007-04-27 15:52:20 UTC (rev 8279)
+++ trunk/cdk-taverna/src/main/java/org/openscience/cdk/applications/taverna/qsar/model/weka/WFilterUnSupervisedResample.java 2007-04-30 12:43:18 UTC (rev 8280)
@@ -26,6 +26,7 @@
package org.openscience.cdk.applications.taverna.qsar.model.weka;
import java.util.HashMap;
+import java.util.Iterator;
import java.util.Map;
import java.util.ArrayList;
import java.util.List;
@@ -110,10 +111,13 @@
public Map execute(Map inputs) throws TaskExecutionException {
Selection selection = new Selection();
Standardise standardise = new Standardise();
- if (title == null || title.length == 0) title = selection.def(inputs, inputNames[0]);
+ if (title == null || title.length == 0) title = selection.def(inputs);
+ else
+ selection.setDescriptor(title);
if (cases == null) cases = standardise.def(title);
standardise.setCases(cases);
- Map newInputs = selection.selection(title, inputNames[0], inputs);
+ standardise.setDescriptor(title);
+ Map newInputs = selection.execute(inputs);
if (y == null){
y = new String[(Integer)newInputs.get("moleculeCount")];
for (int i = 0; i < (Integer)newInputs.get("moleculeCount"); i++){
@@ -121,7 +125,19 @@
}
}
if (classAttrib == null) classAttrib = new String[]{"unknown"};
- Double[][] matrix = standardise.standardise(title, newInputs, (Integer)((newInputs.get("moleculeCount"))));
+ List matrixtmp = (List)((DataThing)(standardise.execute(newInputs)).get("matrix")).getDataObject();
+ Double[][] matrix = new Double[matrixtmp.size()][((List)matrixtmp.get(0)).size()];
+ int z = 0 , s = 0;
+ for (Iterator iter = matrixtmp.iterator(); iter.hasNext();) {
+ List element = (List) iter.next();
+ for (Iterator iterator = element.iterator(); iterator.hasNext();) {
+ Double val = (Double) iterator.next();
+ matrix[z][s] = val;
+ s++;
+ }
+ z++;
+ s = 0;
+ }
String[] attrib = new String[title.length];
for (int i = 0; i < title.length; i++) {
attrib[i] = title[i].substring(title[i].lastIndexOf(".") + 1);
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|