/** * Adds a feature for each example on to the current feature vector * * @param toAdd the feature vector to add */ @Override public void addFeatureVector(INDArray toAdd) { setFeatures(Nd4j.hstack(getFeatureMatrix(), toAdd)); }
switch (features.rank()) { case 2: first.setFeatures(features.get(interval(0, numHoldout), all())); second.setFeatures(features.get(interval(numHoldout, numExamples), all())); break; case 3: first.setFeatures(features.get(interval(0, numHoldout), all(), all())); second.setFeatures(features.get(interval(numHoldout, numExamples), all(), all())); break; case 4: first.setFeatures(features.get(interval(0, numHoldout), all(), all(), all())); second.setFeatures(features.get(interval(numHoldout, numExamples), all(), all(), all())); break; default:
/** * Transform the data * @param dataSet the dataset to transform */ public void transform(DataSet dataSet) { dataSet.setFeatures(dataSet.getFeatures().subRowVector(mean)); dataSet.setFeatures(dataSet.getFeatures().divRowVector(std)); }
switch (features.rank()) { case 2: first.setFeatures(features.get(interval(0, numHoldout), all())); second.setFeatures(features.get(interval(numHoldout, numExamples), all())); break; case 3: first.setFeatures(features.get(interval(0, numHoldout), all(), all())); second.setFeatures(features.get(interval(numHoldout, numExamples), all(), all())); break; case 4: first.setFeatures(features.get(interval(0, numHoldout), all(), all(), all())); second.setFeatures(features.get(interval(numHoldout, numExamples), all(), all(), all())); break; default:
/** * @Deprecated * Subtract by the column means and divide by the standard deviation */ @Deprecated @Override public void normalizeZeroMeanZeroUnitVariance() { INDArray columnMeans = getFeatures().mean(0); INDArray columnStds = getFeatureMatrix().std(0); setFeatures(getFeatures().subiRowVector(columnMeans)); columnStds.addi(Nd4j.scalar(Nd4j.EPS_THRESHOLD)); setFeatures(getFeatures().diviRowVector(columnStds)); }
setFeatures(filtered.getFeatures()); setLabels(newLabelMatrix);
/** * Transform the data * @param dataSet the dataset to transform */ public void transform(DataSet dataSet) { dataSet.setFeatures(dataSet.getFeatures().subRowVector(mean)); dataSet.setFeatures(dataSet.getFeatures().divRowVector(std)); }
/** * Adds a feature for each example on to the current feature vector * * @param toAdd the feature vector to add */ @Override public void addFeatureVector(INDArray toAdd) { setFeatures(Nd4j.hstack(getFeatureMatrix(), toAdd)); }
/** * @Deprecated * Subtract by the column means and divide by the standard deviation */ @Deprecated @Override public void normalizeZeroMeanZeroUnitVariance() { INDArray columnMeans = getFeatures().mean(0); INDArray columnStds = getFeatureMatrix().std(0); setFeatures(getFeatures().subiRowVector(columnMeans)); columnStds.addi(Nd4j.scalar(Nd4j.EPS_THRESHOLD)); setFeatures(getFeatures().diviRowVector(columnStds)); }
newFeatures.tensorAlongDimension(1, new int[] {0, 2, 3}).subi(uMean).divi(uStd); newFeatures.tensorAlongDimension(2, new int[] {0, 2, 3}).subi(vMean).divi(vStd); result.get(i).setFeatures(newFeatures);
setFeatures(filtered.getFeatures()); setLabels(newLabelMatrix);
vStd += varManual(vChannel, vTempMean); vMean += vTempMean; data.setFeatures(data.getFeatureMatrix().div(255)); } else { data.setFeatures(data.getFeatureMatrix().div(255));