Examples with TetradVector edu.cmu.tetrad.util.TetradVector used on opensource projects

Example 26 with TetradVector

use of edu.cmu.tetrad.util.TetradVector in project tetrad by cmu-phil.

the class Ling method removeZeroRowsAndCols.

private TetradMatrix removeZeroRowsAndCols(TetradMatrix w, List<Node> variables) {
    TetradMatrix _W = w.copy();
    List<Node> _variables = new ArrayList<>(variables);
    List<Integer> remove = new ArrayList<>();
    ROW: for (int i = 0; i < _W.rows(); i++) {
        TetradVector row = _W.getRow(i);
        for (int j = 0; j < row.size(); j++) {
            if (row.get(j) != 0)
                continue ROW;
        }
        remove.add(i);
    }
    COLUMN: for (int i = 0; i < _W.rows(); i++) {
        TetradVector col = _W.getColumn(i);
        for (int j = 0; j < col.size(); j++) {
            if (col.get(j) != 0)
                continue COLUMN;
        }
        if (!remove.contains((i))) {
            remove.add(i);
        }
    }
    int[] rows = new int[_W.rows() - remove.size()];
    int count = -1;
    for (int k = 0; k < w.rows(); k++) {
        if (remove.contains(k)) {
            variables.remove(_variables.get(k));
        } else {
            if (!remove.contains(k))
                rows[++count] = k;
        }
    }
    w = w.getSelection(rows, rows);
    return w;
}

Example 27 with TetradVector

use of edu.cmu.tetrad.util.TetradVector in project tetrad by cmu-phil.

the class Ling method permuteRows.

private static TetradMatrix permuteRows(TetradMatrix mat, List<Integer> permutation) {
    TetradMatrix permutedMat = mat.like();
    for (int j = 0; j < mat.rows(); j++) {
        TetradVector row = mat.getRow(j);
        permutedMat.assignRow(permutation.get(j), row);
    }
    return permutedMat;
}

Example 28 with TetradVector

use of edu.cmu.tetrad.util.TetradVector in project tetrad by cmu-phil.

the class Ling method makeDataSet.

private void makeDataSet(GraphWithParameters graphWP) {
    // define the "Gaussian-squared" distribution
    Distribution gp2 = new GaussianPower(2);
    // the coefficients of the error terms  (here, all 1s)
    TetradVector errorCoefficients = getErrorCoeffsIdentity(graphWP.getGraph().getNumNodes());
    // generate data from the SEM
    TetradMatrix inVectors = simulateCyclic(graphWP, errorCoefficients, numSamples, gp2);
    // reformat it
    dataSet = ColtDataSet.makeContinuousData(graphWP.getGraph().getNodes(), new TetradMatrix(inVectors.transpose().toArray()));
}

Example 29 with TetradVector

use of edu.cmu.tetrad.util.TetradVector in project tetrad by cmu-phil.

the class LingamPattern method getScore.

// =============================PRIVATE METHODS=========================//
private Score getScore(Graph dag, TetradMatrix data, List<Node> variables) {
    // System.out.println("Scoring DAG: " + dag);
    Regression regression = new RegressionDataset(data, variables);
    List<Node> nodes = dag.getNodes();
    double score = 0.0;
    double[] pValues = new double[nodes.size()];
    TetradMatrix residuals = new TetradMatrix(data.rows(), data.columns());
    for (int i = 0; i < nodes.size(); i++) {
        Node _target = nodes.get(i);
        List<Node> _regressors = dag.getParents(_target);
        Node target = getVariable(variables, _target.getName());
        List<Node> regressors = new ArrayList<>();
        for (Node _regressor : _regressors) {
            Node variable = getVariable(variables, _regressor.getName());
            regressors.add(variable);
        }
        RegressionResult result = regression.regress(target, regressors);
        TetradVector residualsColumn = result.getResiduals();
        // residuals.viewColumn(i).assign(residualsColumn);
        residuals.assignColumn(i, residualsColumn);
        DoubleArrayList residualsArray = new DoubleArrayList(residualsColumn.toArray());
        double mean = Descriptive.mean(residualsArray);
        double std = Descriptive.standardDeviation(Descriptive.variance(residualsArray.size(), Descriptive.sum(residualsArray), Descriptive.sumOfSquares(residualsArray)));
        for (int i2 = 0; i2 < residualsArray.size(); i2++) {
            residualsArray.set(i2, (residualsArray.get(i2) - mean) / std);
            residualsArray.set(i2, Math.abs(residualsArray.get(i2)));
        }
        double _mean = Descriptive.mean(residualsArray);
        double diff = _mean - Math.sqrt(2.0 / Math.PI);
        score += diff * diff;
    }
    for (int j = 0; j < residuals.columns(); j++) {
        double[] x = residuals.getColumn(j).toArray();
        double p = new AndersonDarlingTest(x).getP();
        pValues[j] = p;
    }
    return new Score(score, pValues);
}

Example 30 with TetradVector

use of edu.cmu.tetrad.util.TetradVector in project tetrad by cmu-phil.

the class MNLRLikelihood method getLik.

public double getLik(int child_index, int[] parents) {
    double lik = 0;
    Node c = variables.get(child_index);
    List<ContinuousVariable> continuous_parents = new ArrayList<>();
    List<DiscreteVariable> discrete_parents = new ArrayList<>();
    for (int p : parents) {
        Node parent = variables.get(p);
        if (parent instanceof ContinuousVariable) {
            continuous_parents.add((ContinuousVariable) parent);
        } else {
            discrete_parents.add((DiscreteVariable) parent);
        }
    }
    int p = continuous_parents.size();
    List<List<Integer>> cells = adTree.getCellLeaves(discrete_parents);
    // List<List<Integer>> cells = partition(discrete_parents);
    int[] continuousCols = new int[p];
    for (int j = 0; j < p; j++) continuousCols[j] = nodesHash.get(continuous_parents.get(j));
    for (List<Integer> cell : cells) {
        int r = cell.size();
        if (r > 1) {
            double[] mean = new double[p];
            double[] var = new double[p];
            for (int i = 0; i < p; i++) {
                for (int j = 0; j < r; j++) {
                    mean[i] += continuousData[continuousCols[i]][cell.get(j)];
                    var[i] += Math.pow(continuousData[continuousCols[i]][cell.get(j)], 2);
                }
                mean[i] /= r;
                var[i] /= r;
                var[i] -= Math.pow(mean[i], 2);
                var[i] = Math.sqrt(var[i]);
                if (Double.isNaN(var[i])) {
                    System.out.println(var[i]);
                }
            }
            int degree = fDegree;
            if (fDegree < 1) {
                degree = (int) Math.floor(Math.log(r));
            }
            TetradMatrix subset = new TetradMatrix(r, p * degree + 1);
            for (int i = 0; i < r; i++) {
                subset.set(i, p * degree, 1);
                for (int j = 0; j < p; j++) {
                    for (int d = 0; d < degree; d++) {
                        subset.set(i, p * d + j, Math.pow((continuousData[continuousCols[j]][cell.get(i)] - mean[j]) / var[j], d + 1));
                    }
                }
            }
            if (c instanceof ContinuousVariable) {
                TetradVector target = new TetradVector(r);
                for (int i = 0; i < r; i++) {
                    target.set(i, continuousData[child_index][cell.get(i)]);
                }
                lik += multipleRegression(target, subset);
            } else {
                ArrayList<Integer> temp = new ArrayList<>();
                TetradMatrix target = new TetradMatrix(r, ((DiscreteVariable) c).getNumCategories());
                for (int i = 0; i < r; i++) {
                    for (int j = 0; j < ((DiscreteVariable) c).getNumCategories(); j++) {
                        target.set(i, j, -1);
                    }
                    target.set(i, discreteData[child_index][cell.get(i)], 1);
                }
                lik += MultinomialLogisticRegression(target, subset);
            }
        }
    }
    return lik;
}