/** {@inheritDoc} */
@Override
public Object read(final InputStream is) {
final BoltzmannMachine result = new BoltzmannMachine();
final EncogReadHelper in = new EncogReadHelper(is);
EncogFileSection section;
while ((section = in.readNextSection()) != null) {
if (section.getSectionName().equals("BOLTZMANN")
&& section.getSubSectionName().equals("PARAMS")) {
final Map<String, String> params = section.parseParams();
result.getProperties().putAll(params);
}
if (section.getSectionName().equals("BOLTZMANN")
&& section.getSubSectionName().equals("NETWORK")) {
final Map<String, String> params = section.parseParams();
result.setWeights(
NumberList.fromList(CSVFormat.EG_FORMAT, params.get(PersistConst.WEIGHTS)));
result.setCurrentState(
NumberList.fromList(CSVFormat.EG_FORMAT, params.get(PersistConst.OUTPUT)));
result.setNeuronCount(EncogFileSection.parseInt(params, PersistConst.NEURON_COUNT));
result.setThreshold(
NumberList.fromList(CSVFormat.EG_FORMAT, params.get(PersistConst.THRESHOLDS)));
result.setAnnealCycles(EncogFileSection.parseInt(params, BoltzmannMachine.ANNEAL_CYCLES));
result.setRunCycles(EncogFileSection.parseInt(params, BoltzmannMachine.RUN_CYCLES));
result.setTemperature(EncogFileSection.parseDouble(params, PersistConst.TEMPERATURE));
}
}
return result;
}
/**
* Create a LMA trainer.
*
* @param method The method to use.
* @param training The training data to use.
* @param argsStr The arguments to use.
* @return The newly created trainer.
*/
public MLTrain create(final MLMethod method, final MLDataSet training, final String argsStr) {
if (!(method instanceof SOM)) {
throw new EncogError(
"Neighborhood training cannot be used on a method of type: "
+ method.getClass().getName());
}
final Map<String, String> args = ArchitectureParse.parseParams(argsStr);
final ParamsHolder holder = new ParamsHolder(args);
final double learningRate = holder.getDouble(MLTrainFactory.PROPERTY_LEARNING_RATE, false, 0.7);
final String neighborhoodStr =
holder.getString(MLTrainFactory.PROPERTY_NEIGHBORHOOD, false, "rbf");
final String rbfTypeStr = holder.getString(MLTrainFactory.PROPERTY_RBF_TYPE, false, "gaussian");
RBFEnum t;
if (rbfTypeStr.equalsIgnoreCase("Gaussian")) {
t = RBFEnum.Gaussian;
} else if (rbfTypeStr.equalsIgnoreCase("Multiquadric")) {
t = RBFEnum.Multiquadric;
} else if (rbfTypeStr.equalsIgnoreCase("InverseMultiquadric")) {
t = RBFEnum.InverseMultiquadric;
} else if (rbfTypeStr.equalsIgnoreCase("MexicanHat")) {
t = RBFEnum.MexicanHat;
} else {
t = RBFEnum.Gaussian;
}
NeighborhoodFunction nf = null;
if (neighborhoodStr.equalsIgnoreCase("bubble")) {
nf = new NeighborhoodBubble(1);
} else if (neighborhoodStr.equalsIgnoreCase("rbf")) {
final String str = holder.getString(MLTrainFactory.PROPERTY_DIMENSIONS, true, null);
final int[] size = NumberList.fromListInt(CSVFormat.EG_FORMAT, str);
nf = new NeighborhoodRBF(size, t);
} else if (neighborhoodStr.equalsIgnoreCase("rbf1d")) {
nf = new NeighborhoodRBF1D(t);
}
if (neighborhoodStr.equalsIgnoreCase("single")) {
nf = new NeighborhoodSingle();
}
final BasicTrainSOM result = new BasicTrainSOM((SOM) method, learningRate, training, nf);
if (args.containsKey(MLTrainFactory.PROPERTY_ITERATIONS)) {
final int plannedIterations = holder.getInt(MLTrainFactory.PROPERTY_ITERATIONS, false, 1000);
final double startRate =
holder.getDouble(MLTrainFactory.PROPERTY_START_LEARNING_RATE, false, 0.05);
final double endRate =
holder.getDouble(MLTrainFactory.PROPERTY_END_LEARNING_RATE, false, 0.05);
final double startRadius = holder.getDouble(MLTrainFactory.PROPERTY_START_RADIUS, false, 10);
final double endRadius = holder.getDouble(MLTrainFactory.PROPERTY_END_RADIUS, false, 1);
result.setAutoDecay(plannedIterations, startRate, endRate, startRadius, endRadius);
}
return result;
}
/**
* Normalize the input file. Write to the specified file.
*
* @param file The file to write to.
*/
public void normalize(final File file) {
if (this.analyst == null) {
throw new EncogError("Can't normalize yet, file has not been analyzed.");
}
ReadCSV csv = null;
PrintWriter tw = null;
try {
csv = new ReadCSV(getInputFilename().toString(), isExpectInputHeaders(), getFormat());
tw = new PrintWriter(new FileWriter(file));
// write headers, if needed
if (isProduceOutputHeaders()) {
writeHeaders(tw);
}
resetStatus();
final int outputLength = this.analyst.determineTotalColumns();
// write file contents
while (csv.next() && !shouldStop()) {
updateStatus(false);
double[] output =
AnalystNormalizeCSV.extractFields(
this.analyst, this.analystHeaders, csv, outputLength, false);
if (this.series.getTotalDepth() > 1) {
output = this.series.process(output);
}
if (output != null) {
final StringBuilder line = new StringBuilder();
NumberList.toList(getFormat(), line, output);
tw.println(line);
}
}
} catch (final IOException e) {
throw new QuantError(e);
} finally {
reportDone(false);
if (csv != null) {
try {
csv.close();
} catch (final Exception ex) {
EncogLogging.log(ex);
}
}
if (tw != null) {
try {
tw.close();
} catch (final Exception ex) {
EncogLogging.log(ex);
}
}
}
}
/** {@inheritDoc} */
@Override
public ActivationFunction createActivationFunction(String fn) {
String name;
double[] params;
int index = fn.indexOf('[');
if (index != -1) {
name = fn.substring(0, index).toLowerCase();
int index2 = fn.indexOf(']');
if (index2 == -1) {
throw new EncogError("Unbounded [ while parsing activation function.");
}
String a = fn.substring(index + 1, index2);
params = NumberList.fromList(CSVFormat.EG_FORMAT, a);
} else {
name = fn.toLowerCase();
params = new double[0];
}
ActivationFunction af = allocateAF(name);
if (af == null) {
return null;
}
if (af.getParamNames().length != params.length) {
throw new EncogError(
name
+ " expected "
+ af.getParamNames().length
+ ", but "
+ params.length
+ " were provided.");
}
for (int i = 0; i < af.getParamNames().length; i++) {
af.setParam(i, params[i]);
}
return af;
}
