private CaseExpanderWrapper(DataWrapper wrapper) {
DataFilter filter = new CaseExpander();
DataSet columnDataModel = (DataSet) wrapper.getSelectedDataModel();
setDataModel(filter.filter(columnDataModel));
setSourceGraph(wrapper.getSourceGraph());
LogDataUtils.logDataModelList(
"Data in which case multipliers of parent node data have been multiplied out.",
getDataModelList());
}
/**
* Constructs the wrapper given some data and the params.
*
* @param data
* @param params
*/
public SplitCasesWrapper(DataWrapper data, SplitCasesParams params) {
if (data == null) {
throw new NullPointerException("The given data must not be null");
}
if (params == null) {
throw new NullPointerException("The given parameters must not be null");
}
DataSet originalData = (DataSet) data.getSelectedDataModel();
DataModel model = createSplits(originalData, params);
this.setDataModel(model);
this.setSourceGraph(data.getSourceGraph());
LogDataUtils.logDataModelList("One split of the parent data.", getDataModelList());
}
public RegressionInterpolatorWrapper(DataWrapper data) {
if (data == null) {
throw new NullPointerException("The givan data must not be null");
}
DataModel model = data.getSelectedDataModel();
if ((!(model instanceof DataSet))) {
throw new IllegalArgumentException("Data must be tabular");
}
DataFilter interpolator = new RegressionInterpolator();
this.setDataModel(interpolator.filter((DataSet) model));
this.setSourceGraph(data.getSourceGraph());
LogDataUtils.logDataModelList(
"Parent data in which missing values have been replaced by regression predictions.",
getDataModelList());
}
public CopyAllDatasetsWrapper(DataWrapper wrapper) {
LogDataUtils.logDataModelList(
"Parent data in which constant columns have been removed.", getDataModelList());
DataModelList inList = wrapper.getDataModelList();
DataModelList outList = new DataModelList();
for (DataModel model : inList) {
if (!(model instanceof DataSet)) {
throw new IllegalArgumentException("Not a data set: " + model.getName());
}
this.setDataModel(model);
outList.add(model);
}
setDataModel(outList);
setSourceGraph(wrapper.getSourceGraph());
}
public ImpliedCovarianceDataAllWrapper(SemEstimatorWrapper wrapper) {
// int sampleSize = params.getSampleSize();
// boolean latentDataSaved = params.isIncludeLatents();
SemEstimator semEstimator = wrapper.getSemEstimator();
SemIm semIm1 = semEstimator.getEstimatedSem();
if (semIm1 != null) {
TetradMatrix matrix2D = semIm1.getImplCovar(true);
int sampleSize = semIm1.getSampleSize();
List<Node> variables =
wrapper.getSemEstimator().getEstimatedSem().getSemPm().getVariableNodes();
CovarianceMatrix cov = new CovarianceMatrix(variables, matrix2D, sampleSize);
setDataModel(cov);
setSourceGraph(wrapper.getSemEstimator().getEstimatedSem().getSemPm().getGraph());
this.semIm = wrapper.getEstimatedSemIm();
}
LogDataUtils.logDataModelList(
"Data simulated from a linear structural equation model.", getDataModelList());
}
/**
* Splits the given data set by collinear columns.
*
* @param wrapper
*/
public SimulateFromCovWrapper(DataWrapper wrapper) {
if (wrapper == null) {
throw new NullPointerException("The given data must not be null");
}
DataModel model = wrapper.getSelectedDataModel();
if (model instanceof ICovarianceMatrix) {
CovarianceMatrix covarianceMatrix = new CovarianceMatrix((CovarianceMatrix) model);
DataSet dataSet = DataUtils.choleskySimulation(covarianceMatrix);
setDataModel(dataSet);
setSourceGraph(wrapper.getSourceGraph());
} else {
throw new IllegalArgumentException("Must be a dataset or a covariance matrix");
}
LogDataUtils.logDataModelList(
"Conversion of data to covariance matrix form.", getDataModelList());
}
