/**
* Copyright (C) 2001-2017 by RapidMiner and the contributors
*
* Complete list of developers available at our web site:
*
* http://rapidminer.com
*
* This program is free software: you can redistribute it and/or modify it under the terms of the
* GNU Affero General Public License as published by the Free Software Foundation, either version 3
* of the License, or (at your option) any later version.
*
* This program is distributed in the hope that it will be useful, but WITHOUT ANY WARRANTY; without
* even the implied warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE. See the GNU
* Affero General Public License for more details.
*
* You should have received a copy of the GNU Affero General Public License along with this program.
* If not, see http://www.gnu.org/licenses/.
*/
package com.rapidminer.operator;
import java.util.Arrays;
import java.util.Iterator;
import java.util.List;
import com.rapidminer.example.ExampleSet;
import com.rapidminer.operator.ports.InputPort;
import com.rapidminer.operator.ports.OutputPort;
import com.rapidminer.operator.ports.metadata.ExampleSetMetaData;
import com.rapidminer.operator.ports.metadata.ModelApplicationRule;
import com.rapidminer.operator.ports.metadata.ModelMetaData;
import com.rapidminer.operator.ports.metadata.PassThroughRule;
import com.rapidminer.operator.ports.metadata.SimplePrecondition;
import com.rapidminer.operator.preprocessing.PreprocessingOperator;
import com.rapidminer.parameter.ParameterType;
import com.rapidminer.parameter.ParameterTypeBoolean;
import com.rapidminer.parameter.ParameterTypeList;
import com.rapidminer.parameter.ParameterTypeString;
/**
* This operator applies a {@link Model} to an {@link ExampleSet}. All parameters of the training
* process should be stored within the model. However, this operator is able to take any parameters
* for the rare case that the particular model evaluates parameters during application. Models can
* be read from a file by using a {@link com.rapidminer.extension.legacy.operator.io.ModelLoader}.
*
* @author Ingo Mierswa, Simon Fischer
*/
public class ModelApplier extends Operator {
/** The parameter name for "key" */
public static final String PARAMETER_KEY = "key";
/** The parameter name for "value" */
public static final String PARAMETER_VALUE = "value";
/** The possible parameters used by the model during application time. */
public static final String PARAMETER_APPLICATION_PARAMETERS = "application_parameters";
/** Indicates if preprocessing models should create a view instead of changing the data. */
private static final String PARAMETER_CREATE_VIEW = "create_view";
/** Last version to silently log unsupported parameters. */
private static final OperatorVersion VERSION_ERROR_UNSUPPORTED_PARAMETER = new OperatorVersion(7, 1, 1);
@Override
public OperatorVersion[] getIncompatibleVersionChanges() {
OperatorVersion[] changes = super.getIncompatibleVersionChanges();
changes = Arrays.copyOf(changes, changes.length + 1);
changes[changes.length - 1] = VERSION_ERROR_UNSUPPORTED_PARAMETER;
return changes;
}
private final InputPort modelInput = getInputPorts().createPort("model");
private final InputPort exampleSetInput = getInputPorts().createPort("unlabelled data");
private final OutputPort exampleSetOutput = getOutputPorts().createPort("labelled data");
private final OutputPort modelOutput = getOutputPorts().createPort("model");
public ModelApplier(OperatorDescription description) {
super(description);
modelInput.addPrecondition(
new SimplePrecondition(modelInput, new ModelMetaData(Model.class, new ExampleSetMetaData())));
exampleSetInput.addPrecondition(new SimplePrecondition(exampleSetInput, new ExampleSetMetaData()));
getTransformer().addRule(new ModelApplicationRule(exampleSetInput, exampleSetOutput, modelInput, false));
getTransformer().addRule(new PassThroughRule(modelInput, modelOutput, false));
}
/**
* Applies the operator and labels the {@link ExampleSet}. The example set in the input is not
* consumed.
*/
@Override
public void doWork() throws OperatorException {
ExampleSet inputExampleSet = exampleSetInput.getData(ExampleSet.class);
Model model = modelInput.getData(Model.class);
if (AbstractModel.class.isAssignableFrom(model.getClass())) {
((AbstractModel) model).setOperator(this);
((AbstractModel) model).setShowProgress(true);
}
log("Set parameters for " + model.getClass().getName());
List<String[]> modelParameters = getParameterList(PARAMETER_APPLICATION_PARAMETERS);
Iterator<String[]> i = modelParameters.iterator();
while (i.hasNext()) {
String[] parameter = i.next();
try {
model.setParameter(parameter[0], parameter[1]);
} catch (UnsupportedApplicationParameterError e) {
if (getCompatibilityLevel().isAtMost(VERSION_ERROR_UNSUPPORTED_PARAMETER)) {
log("The learned model does not support parameter");
} else {
e.setOperator(this);
throw e;
}
}
}
// handling PreprocessingModels: extra treatment for views
if (getParameterAsBoolean(PARAMETER_CREATE_VIEW)) {
try {
model.setParameter(PreprocessingOperator.PARAMETER_CREATE_VIEW, true);
} catch (UnsupportedApplicationParameterError e) {
if (getCompatibilityLevel().isAtMost(VERSION_ERROR_UNSUPPORTED_PARAMETER)) {
log("The learned model does not have a view to create");
} else {
e.setOperator(this);
throw e;
}
}
}
log("Applying " + model.getClass().getName());
ExampleSet result = inputExampleSet;
try {
result = model.apply(inputExampleSet);
} catch (UserError e) {
if (e.getOperator() == null) {
e.setOperator(this);
}
throw e;
}
if (AbstractModel.class.isAssignableFrom(model.getClass())) {
((AbstractModel) model).setOperator(null);
((AbstractModel) model).setShowProgress(false);
}
exampleSetOutput.deliver(result);
modelOutput.deliver(model);
}
@Override
public boolean shouldAutoConnect(OutputPort port) {
if (port == modelOutput) {
return getParameterAsBoolean("keep_model");
} else {
return super.shouldAutoConnect(port);
}
}
@Override
public List<ParameterType> getParameterTypes() {
List<ParameterType> types = super.getParameterTypes();
types.add(new ParameterTypeList(PARAMETER_APPLICATION_PARAMETERS,
"Model parameters for application (usually not needed).",
new ParameterTypeString(PARAMETER_KEY, "The model parameter key."),
new ParameterTypeString(PARAMETER_VALUE, "This key's value")));
types.add(new ParameterTypeBoolean(PARAMETER_CREATE_VIEW,
"Indicates that models should create a new view on the data where possible. Then, instead of changing the data itself, the results are calculated on the fly if needed.",
false));
return types;
}
}