public void testClassification() throws Exception {
File rawFile = TEMP_DIR.createFile("simple.csv");
File egaFile = TEMP_DIR.createFile("simple.ega");
File outputFile = TEMP_DIR.createFile("simple_output.csv");
FileUtil.copyResource("org/encog/data/simple.csv", rawFile);
FileUtil.copyResource("org/encog/data/simple-r.ega", egaFile);
EncogAnalyst analyst = new EncogAnalyst();
analyst.addAnalystListener(new ConsoleAnalystListener());
analyst.load(egaFile);
analyst.executeTask("task-full");
ReadCSV csv = new ReadCSV(outputFile.toString(), true, CSVFormat.ENGLISH);
while (csv.next()) {
double diff = Math.abs(csv.getDouble(2) - csv.getDouble(4));
Assert.assertTrue(diff < 1.5);
}
Assert.assertEquals(4, analyst.getScript().getFields().length);
Assert.assertEquals(3, analyst.getScript().getFields()[3].getClassMembers().size());
csv.close();
}
/**
* Load a CSV file into a memory dataset.
*
* @param format The CSV format to use.
* @param filename The filename to load.
* @param headers True if there is a header line.
* @param inputSize The input size. Input always comes first in a file.
* @param idealSize The ideal size, 0 for unsupervised.
* @return A NeuralDataSet that holds the contents of the CSV file.
*/
public static MLDataSet loadCSVTOMemory(
CSVFormat format, String filename, boolean headers, int inputSize, int idealSize) {
MLDataSet result = new BasicMLDataSet();
ReadCSV csv = new ReadCSV(filename, headers, format);
while (csv.next()) {
MLData input = null;
MLData ideal = null;
int index = 0;
input = new BasicMLData(inputSize);
for (int i = 0; i < inputSize; i++) {
double d = csv.getDouble(index++);
input.setData(i, d);
}
if (idealSize > 0) {
ideal = new BasicMLData(idealSize);
for (int i = 0; i < idealSize; i++) {
double d = csv.getDouble(index++);
ideal.setData(i, d);
}
}
MLDataPair pair = new BasicMLDataPair(input, ideal);
result.add(pair);
}
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);
}
}
}
}
public static void convertCSV2Binary(
File csvFile, CSVFormat format, File binFile, int[] input, int[] ideal, boolean headers) {
binFile.delete();
ReadCSV csv = new ReadCSV(csvFile.toString(), headers, format);
BufferedMLDataSet buffer = new BufferedMLDataSet(binFile);
buffer.beginLoad(input.length, ideal.length);
while (csv.next()) {
BasicMLData inputData = new BasicMLData(input.length);
BasicMLData idealData = new BasicMLData(ideal.length);
// handle input data
for (int i = 0; i < input.length; i++) {
inputData.setData(i, csv.getDouble(input[i]));
}
// handle input data
for (int i = 0; i < ideal.length; i++) {
idealData.setData(i, csv.getDouble(ideal[i]));
}
// add to dataset
buffer.add(inputData, idealData);
}
buffer.endLoad();
}
/**
* Extract fields from a file into a numeric array for machine learning.
*
* @param analyst The analyst to use.
* @param headers The headers for the input data.
* @param csv The CSV that holds the input data.
* @param outputLength The length of the returned array.
* @param skipOutput True if the output should be skipped.
* @return The encoded data.
*/
public static final double[] extractFields(
final EncogAnalyst analyst,
final CSVHeaders headers,
final ReadCSV csv,
final int outputLength,
final boolean skipOutput) {
final double[] output = new double[outputLength];
int outputIndex = 0;
for (final AnalystField stat : analyst.getScript().getNormalize().getNormalizedFields()) {
stat.init();
if (stat.getAction() == NormalizationAction.Ignore) {
continue;
}
if (stat.isOutput() && skipOutput) {
continue;
}
int index = headers.find(stat.getName());
final String str = csv.get(index).trim();
// is this an unknown value?
if (str.equals("?") || str.length() == 0) {
HandleMissingValues handler = analyst.getScript().getNormalize().getMissingValues();
double[] d = handler.handleMissing(analyst, stat);
// should we skip the entire row
if (d == null) {
return null;
}
// copy the returned values in place of the missing values
for (int i = 0; i < d.length; i++) {
output[outputIndex++] = d[i];
}
} else {
// known value
if (stat.getAction() == NormalizationAction.Normalize) {
double d = csv.getFormat().parse(str);
d = stat.normalize(d);
output[outputIndex++] = d;
} else if (stat.getAction() == NormalizationAction.PassThrough) {
double d = csv.getFormat().parse(str);
output[outputIndex++] = d;
} else {
final double[] d = stat.encode(str);
for (final double element : d) {
output[outputIndex++] = element;
}
}
}
}
return output;
}
private void analyzeFile() {
ScriptProperties prop = this.analyst.getScript().getProperties();
// get filenames, headers & format
String sourceID = prop.getPropertyString(ScriptProperties.HEADER_DATASOURCE_RAW_FILE);
File sourceFile = this.analyst.getScript().resolveFilename(sourceID);
CSVFormat inputFormat = this.analyst.getScript().determineFormat();
boolean headers = this.analyst.getScript().expectInputHeaders(sourceID);
// read the file
this.rowCount = 0;
this.missingCount = 0;
ReadCSV csv = new ReadCSV(sourceFile.toString(), headers, inputFormat);
while (csv.next()) {
rowCount++;
if (csv.hasMissing()) missingCount++;
}
csv.close();
}
/**
* Analyze the data. This counts the records and prepares the data to be processed.
*
* @param theAnalyst The analyst to use.
* @param inputFile The input file to analyze.
* @param headers True, if the input file has headers.
* @param format The format of the input file.
*/
public void analyze(
final EncogAnalyst theAnalyst,
final File inputFile,
final boolean headers,
final CSVFormat format) {
this.setInputFilename(inputFile);
setExpectInputHeaders(headers);
setInputFormat(format);
setAnalyzed(true);
this.analyst = theAnalyst;
this.data = new BasicMLDataSet();
resetStatus();
int recordCount = 0;
final int outputLength = this.analyst.determineTotalColumns();
final ReadCSV csv =
new ReadCSV(
this.getInputFilename().toString(), this.isExpectInputHeaders(), this.getFormat());
readHeaders(csv);
this.analystHeaders = new CSVHeaders(this.getInputHeadings());
while (csv.next() && !shouldStop()) {
updateStatus(true);
final double[] inputArray =
AnalystNormalizeCSV.extractFields(analyst, this.analystHeaders, csv, outputLength, true);
final MLData input = new BasicMLData(inputArray);
this.data.add(new BasicMLDataPair(input));
recordCount++;
}
setRecordCount(recordCount);
this.setColumnCount(csv.getColumnCount());
readHeaders(csv);
csv.close();
reportDone(true);
}
/**
* Construct the object.
*
* @param filename The filename.
* @param headers False if headers are not extended.
* @param format The CSV format.
*/
public CSVHeaders(final File filename, final boolean headers, final CSVFormat format) {
ReadCSV csv = null;
try {
csv = new ReadCSV(filename.toString(), headers, format);
if (csv.next()) {
if (headers) {
for (final String str : csv.getColumnNames()) {
this.headerList.add(str);
}
} else {
for (int i = 0; i < csv.getColumnCount(); i++) {
this.headerList.add("field:" + (i + 1));
}
}
}
init();
} finally {
if (csv != null) {
csv.close();
}
}
}
