Use Matrix initializer_list constructor instead of comma initializers

doc/Code/LocalizationFactor.cpp

@@ -11,9 +11,7 @@ class UnaryFactor: public NoiseModelFactor1<Pose2> {
 
   Vector evaluateError(const Pose2& q, OptionalMatrixType H) const override {
     const Rot2& R = q.rotation();
-    if (H) (*H) = (gtsam::Matrix(2, 3) <<
-            R.c(), -R.s(), 0.0,
-            R.s(), R.c(), 0.0).finished();
-    return (Vector(2) << q.x() - mx_, q.y() - my_).finished();
+    if (H) (*H) = gtsam::Matrix23{{R.c(), -R.s(), 0.0}, {R.s(), R.c(), 0.0}};
+    return Vector{{q.x() - mx_, q.y() - my_}};
   }
 };

examples/CombinedImuFactorsExample.cpp

@@ -175,8 +175,7 @@ int main(int argc, char* argv[]) {
 
   // Assemble prior noise model and add it the graph.`
   auto pose_noise_model = noiseModel::Diagonal::Sigmas(
-      (Vector(6) << 0.01, 0.01, 0.01, 0.5, 0.5, 0.5)
-          .finished());  // rad,rad,rad,m, m, m
+      Vector{{0.01, 0.01, 0.01, 0.5, 0.5, 0.5}});  // rad,rad,rad,m, m, m
   auto velocity_noise_model = noiseModel::Isotropic::Sigma(3, 0.1);  // m/s
   auto bias_noise_model = noiseModel::Isotropic::Sigma(6, 1e-3);
 

examples/GaussianProcessWnoaSE3Example.cpp

@@ -117,9 +117,9 @@ int main() {
   // Define measurement noise model and WNOA process noise covariance.
   // These define the prior belief about measurement and process uncertainties.
   Eigen::Matrix<double, kDoF, kDoF> measurement_cov =
-      1e-2 * (Eigen::Matrix<double, kDoF, 1>() << 0.1, 0.1, 0.1, 1.0, 1.0, 1.0)
-                 .finished()
-                 .asDiagonal();
+      1e-2 *
+      Eigen::Matrix<double, kDoF, 1>{{0.1}, {0.1}, {0.1}, {1.0}, {1.0}, {1.0}}
+          .asDiagonal();
   const auto noise_model =
       gtsam::noiseModel::Gaussian::Covariance(measurement_cov);
 
@@ -182,8 +182,7 @@ int main() {
 
   // Create initial trajectory estimate using constant velocity assumption.
   // This provides a reasonable starting point for the optimizer.
-  const Velocity initial_twist =
-      (Velocity() << 0.0, 0.0, 0.0, -1.0, 0.0, 0.0).finished();
+  const Velocity initial_twist{{0.0}, {0.0}, {0.0}, {-1.0}, {0.0}, {0.0}};
   gtsam::Values values_init;
   Pose running_pose = values_gt.at<Pose>(estimated_states_vec.front().pose);
   values_init.insert(estimated_states_vec.front().pose, running_pose);

examples/IMUKittiExampleGPS.cpp

@@ -160,11 +160,9 @@ int main(int argc, char* argv[]) {
   vector<GpsMeasurement> gps_measurements;
   loadKittiData(kitti_calibration, imu_measurements, gps_measurements);
 
-  Vector6 BodyP =
-      (Vector6() << kitti_calibration.body_ptx, kitti_calibration.body_pty,
-       kitti_calibration.body_ptz, kitti_calibration.body_prx,
-       kitti_calibration.body_pry, kitti_calibration.body_prz)
-          .finished();
+  Vector6 BodyP{kitti_calibration.body_ptx, kitti_calibration.body_pty,
+                kitti_calibration.body_ptz, kitti_calibration.body_prx,
+                kitti_calibration.body_pry, kitti_calibration.body_prz};
   auto body_T_imu = Pose3::Expmap(BodyP);
   if (!body_T_imu.equals(Pose3(), 1e-5)) {
     printf(

examples/ImuFactorsExample.cpp

@@ -182,8 +182,7 @@ int main(int argc, char* argv[]) {
 
   // Assemble prior noise model and add it the graph.`
   auto pose_noise_model = noiseModel::Diagonal::Sigmas(
-      (Vector(6) << 0.01, 0.01, 0.01, 0.5, 0.5, 0.5)
-          .finished());  // rad,rad,rad,m, m, m
+      Vector{{0.01, 0.01, 0.01, 0.5, 0.5, 0.5}});  // rad,rad,rad,m, m, m
   auto velocity_noise_model = noiseModel::Isotropic::Sigma(3, 0.1);  // m/s
   auto bias_noise_model = noiseModel::Isotropic::Sigma(6, 1e-3);
 

examples/LeggedEstimatorReplayExample.cpp

@@ -71,9 +71,8 @@ struct ReplayConfig : public LeggedEstimatorParams {
   Vector3 gravity = Vector3(0.0, 0.0, -9.81);
   Point3 initialPosition = Point3(0.0, 0.0, 0.76);
   Vector3 initialVelocity = Vector3::Zero();
-  Vector initialBaseCovarianceDiagonal =
-      (Vector(9) << 1e-2, 1e-2, 1e-6, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05)
-          .finished();
+  Vector initialBaseCovarianceDiagonal{
+      {1e-2, 1e-2, 1e-6, 0.05, 0.05, 0.05, 0.05, 0.05, 0.05}};
   double sigmaGyro = 8e-4;
   double sigmaIntegration = 1e-3;
   double sigmaAcc = 2e-2;

examples/LocalizationExample.cpp

@@ -97,8 +97,8 @@ class UnaryFactor: public NoiseModelFactorN<Pose2> {
     // H =  [ cos(q.theta)  -sin(q.theta) 0 ]
     //      [ sin(q.theta)   cos(q.theta) 0 ]
     const Rot2& R = q.rotation();
-    if (H) (*H) = (gtsam::Matrix(2, 3) << R.c(), -R.s(), 0.0, R.s(), R.c(), 0.0).finished();
-    return (Vector(2) << q.x() - mx_, q.y() - my_).finished();
+    if (H) (*H) = gtsam::Matrix{{R.c(), -R.s(), 0.0}, {R.s(), R.c(), 0.0}};
+    return Vector{{q.x() - mx_, q.y() - my_}};
   }
 
   // The second is a 'clone' function that allows the factor to be copied. Under most

examples/Pose2SLAMExample_graph.cpp

@@ -34,7 +34,7 @@ int main (int argc, char** argv) {
   // we are in build/examples, data is in examples/Data
   NonlinearFactorGraph::shared_ptr graph;
   Values::shared_ptr initial;
-  SharedDiagonal model = noiseModel::Diagonal::Sigmas((Vector(3) << 0.05, 0.05, 5.0 * M_PI / 180.0).finished());
+  SharedDiagonal model = noiseModel::Diagonal::Sigmas(Vector{{0.05, 0.05, 5.0 * M_PI / 180.0}});
   string graph_file = findExampleDataFile("w100.graph");
   std::tie(graph, initial) = load2D(graph_file, model);
   initial->print("Initial estimate:\n");

examples/Pose3Localization.cpp

@@ -41,7 +41,7 @@ int main(const int argc, const char* argv[]) {
 
   // Add prior on the first key
   auto priorModel = noiseModel::Diagonal::Variances(
-      (Vector(6) << 1e-6, 1e-6, 1e-6, 1e-4, 1e-4, 1e-4).finished());
+      Vector{{1e-6, 1e-6, 1e-6, 1e-4, 1e-4, 1e-4}});
   Key firstKey = 0;
   for (const auto key : initial->keys()) {
     std::cout << "Adding prior to g2o file " << std::endl;

examples/Pose3SLAMExampleExpressions_BearingRangeWithTransform.cpp

@@ -43,10 +43,10 @@ int main(int argc, char* argv[]) {
   ExpressionFactorGraph graph;
 
   // Specify uncertainty on first pose prior and also for between factor (simplicity reasons)
-  auto poseNoise = noiseModel::Diagonal::Sigmas((Vector(6)<<0.3,0.3,0.3,0.1,0.1,0.1).finished());
+  auto poseNoise = noiseModel::Diagonal::Sigmas(Vector{{0.3,0.3,0.3,0.1,0.1,0.1}});
 
   // Uncertainty bearing range measurement;
-  auto bearingRangeNoise = noiseModel::Diagonal::Sigmas((Vector(3)<<0.01,0.03,0.05).finished());
+  auto bearingRangeNoise = noiseModel::Diagonal::Sigmas(Vector{{0.01,0.03,0.05}});
 
   // Expressions for body-frame at key 0 and sensor-tf
   Pose3_ x_('x', 0);
@@ -99,4 +99,4 @@ int main(int argc, char* argv[]) {
 
   return 0;
 }
-/* ************************************************************************* */
+/* ************************************************************************* */

examples/Pose3SLAMExample_g2o.cpp

@@ -40,7 +40,7 @@ int main(const int argc, const char* argv[]) {
 
   // Add prior on the first key
   auto priorModel = noiseModel::Diagonal::Variances(
-      (Vector(6) << 1e-6, 1e-6, 1e-6, 1e-4, 1e-4, 1e-4).finished());
+      Vector{{1e-6, 1e-6, 1e-6, 1e-4, 1e-4, 1e-4}});
   Key firstKey = 0;
   for (const auto key : initial->keys()) {
     std::cout << "Adding prior to g2o file " << std::endl;

examples/Pose3SLAMExample_initializePose3Chordal.cpp

@@ -40,7 +40,7 @@ int main(const int argc, const char* argv[]) {
 
   // Add prior on the first key
   auto priorModel = noiseModel::Diagonal::Variances(
-      (Vector(6) << 1e-6, 1e-6, 1e-6, 1e-4, 1e-4, 1e-4).finished());
+      Vector{{1e-6, 1e-6, 1e-6, 1e-4, 1e-4, 1e-4}});
   Key firstKey = 0;
   for (const auto key : initial->keys()) {
     std::cout << "Adding prior to g2o file " << std::endl;

examples/Pose3SLAMExample_initializePose3Gradient.cpp

@@ -40,7 +40,7 @@ int main(const int argc, const char* argv[]) {
 
   // Add prior on the first key
   auto priorModel = noiseModel::Diagonal::Variances(
-      (Vector(6) << 1e-6, 1e-6, 1e-6, 1e-4, 1e-4, 1e-4).finished());
+      Vector{{1e-6, 1e-6, 1e-6, 1e-4, 1e-4, 1e-4}});
   Key firstKey = 0;
   for (const auto key : initial->keys()) {
     std::cout << "Adding prior to g2o file " << std::endl;

examples/SelfCalibrationExample.cpp

@@ -80,8 +80,8 @@ int main(int argc, char* argv[]) {
                  pointNoise);  // add directly to graph
 
   // Add a prior on the calibration.
-  auto calNoise = noiseModel::Diagonal::Sigmas(
-      (Vector(5) << 500, 500, 0.1, 100, 100).finished());
+  auto calNoise =
+      noiseModel::Diagonal::Sigmas(Vector{{500, 500, 0.1, 100, 100}});
   graph.addPrior(Symbol('K', 0), K, calNoise);
 
   // Create the initial estimate to the solution
@@ -102,4 +102,3 @@ int main(int argc, char* argv[]) {
 
   return 0;
 }
-

examples/elaboratePoint2KalmanFilter.cpp

@@ -259,7 +259,7 @@ int main() {
 
   // And update using z2 ...
   Point2 z2(2.0, 0.0);
-  SharedDiagonal R2 = noiseModel::Diagonal::Sigmas((gtsam::Vector2() << 0.25, 0.25).finished());
+  SharedDiagonal R2 = noiseModel::Diagonal::Sigmas(gtsam::Vector2{0.25, 0.25});
   PriorFactor<Point2> factor8(X(2), z2, R2);
 
   // Linearize the factor and add it to the linear factor graph

gtsam/base/Matrix.h

@@ -293,7 +293,7 @@ GTSAM_EXPORT Matrix vector_scale(const Matrix& A, const Vector& v, bool inf_mask
 */
 
 inline Matrix3 skewSymmetric(double wx, double wy, double wz) {
-  return (Matrix3() << 0.0, -wz, +wy, +wz, 0.0, -wx, -wy, +wx, 0.0).finished();
+  return Matrix3{{0.0, -wz, +wy}, {+wz, 0.0, -wx}, {-wy, +wx, 0.0}};
 }
 
 template <class Derived>

gtsam/base/tests/testCholesky.cpp

@@ -40,14 +40,14 @@ TEST(cholesky, choleskyPartial0) {
 
 /* ************************************************************************* */
 TEST(cholesky, choleskyPartial) {
-  Matrix ABC = (Matrix(7,7) <<
-                      4.0375,   3.4584,   3.5735,   2.4815,   2.1471,   2.7400,   2.2063,
-                          0.,   4.7267,   3.8423,   2.3624,   2.8091,   2.9579,   2.5914,
-                          0.,       0.,   5.1600,   2.0797,   3.4690,   3.2419,   2.9992,
-                          0.,       0.,       0.,   1.8786,   1.0535,   1.4250,   1.3347,
-                          0.,       0.,       0.,       0.,   3.0788,   2.6283,   2.3791,
-                          0.,       0.,       0.,       0.,       0.,   2.9227,   2.4056,
-                          0.,       0.,       0.,       0.,       0.,       0.,   2.5776).finished();
+  Matrix ABC{
+            {4.0375,   3.4584,   3.5735,   2.4815,   2.1471,   2.7400,   2.2063},
+                {0.,   4.7267,   3.8423,   2.3624,   2.8091,   2.9579,   2.5914},
+                {0.,       0.,   5.1600,   2.0797,   3.4690,   3.2419,   2.9992},
+                {0.,       0.,       0.,   1.8786,   1.0535,   1.4250,   1.3347},
+                {0.,       0.,       0.,       0.,   3.0788,   2.6283,   2.3791},
+                {0.,       0.,       0.,       0.,       0.,   2.9227,   2.4056},
+                {0.,       0.,       0.,       0.,       0.,       0.,   2.5776}};
 
   // Do partial Cholesky on 3 frontal scalar variables
   Matrix RSL(ABC);
@@ -68,9 +68,9 @@ TEST(cholesky, choleskyPartial) {
 
 /* ************************************************************************* */
 TEST(cholesky, BadScalingCholesky) {
-  Matrix A = (Matrix(2,2) <<
-      1e-40, 0.0,
-      0.0, 1.0).finished();
+  Matrix A{
+      {1e-40, 0.0},
+      {0.0, 1.0}};
 
   Matrix R(A.transpose() * A);
   choleskyPartial(R, 2);
@@ -83,9 +83,9 @@ TEST(cholesky, BadScalingCholesky) {
 
 /* ************************************************************************* */
 TEST(cholesky, BadScalingSVD) {
-  Matrix A = (Matrix(2,2) <<
-      1.0, 0.0,
-      0.0, 1e-40).finished();
+  Matrix A{
+      {1.0, 0.0},
+      {0.0, 1e-40}};
 
   Matrix U, V;
   Vector S;