renamed sparse to sparseJacobian_

examples/matlab/PlanarSLAMExample_easy.m

@@ -74,12 +74,14 @@ initialEstimate.print('initial estimate');
 result = graph.optimize_(initialEstimate);
 result.print('final result');
 
-%% Print out the corresponding dense matrix
+%% Get the corresponding dense matrix
 ord = graph.orderingCOLAMD(result);
 gfg = graph.linearize(result,ord);
-denseAb = gfg.denseJacobian
+denseAb = gfg.denseJacobian;
 
-%% Get sparse matrix
-IJS = gfg.sparse();
+%% Get sparse matrix A and RHS b
+IJS = gfg.sparseJacobian_();
 Ab=sparse(IJS(1,:),IJS(2,:),IJS(3,:));
-spy(Ab(:,1:end-1));
+A = Ab(:,1:end-1);
+b = full(Ab(:,end));
+spy(A);

gtsam.h

@@ -110,7 +110,7 @@ class GaussianFactorGraph {
 	void combine(const GaussianFactorGraph& lfg);
   Matrix denseJacobian() const;
   Matrix denseHessian() const;
-	Matrix sparse() const;
+	Matrix sparseJacobian_() const;
 };
 
 class Landmark2 {

gtsam/linear/GaussianFactorGraph.cpp

@@ -141,7 +141,7 @@ namespace gtsam {
 	}
 
 	/* ************************************************************************* */
-	Matrix GaussianFactorGraph::sparse() const {
+	Matrix GaussianFactorGraph::sparseJacobian_() const {
 
 		// call sparseJacobian
 		typedef boost::tuple<size_t, size_t, double> triplet;

gtsam/linear/GaussianFactorGraph.h

@@ -160,7 +160,7 @@ namespace gtsam {
 		 * such that S(i(k),j(k)) = s(k), which can be given to MATLAB's sparse.
 		 * The standard deviations are baked into A and b
 		 */
-		Matrix sparse() const;
+		Matrix sparseJacobian_() const;
 
     /**
      * Return a dense \f$ m \times n \f$ Jacobian matrix, augmented with b