Merge pull request #2168 from SCIInstitute/2bpls

Shape/Scalar Correlation

Author: akenmorris

Libs/Analyze/Analyze.cpp

@@ -14,7 +14,7 @@ namespace shapeworks {
 
 //---------------------------------------------------------------------------
 static json get_eigen_vectors(ParticleShapeStatistics* stats) {
-  auto values = stats->Eigenvectors();
+    auto values = stats->get_eigen_vectors();
 
   std::vector<double> vals;
   for (size_t i = values.cols() - 1, ii = 0; i > 0; i--, ii++) {
@@ -173,8 +173,8 @@ void Analyze::run_offline_analysis(std::string outfile, float range, float steps
   double increment = range / half_steps;
 
   std::vector<double> eigen_vals;
-  for (int i = stats_.Eigenvalues().size() - 1; i > 0; i--) {
-    eigen_vals.push_back(stats_.Eigenvalues()[i]);
+  for (int i = stats_.get_eigen_values().size() - 1; i > 0; i--) {
+    eigen_vals.push_back(stats_.get_eigen_values()[i]);
   }
   double sum = std::accumulate(eigen_vals.begin(), eigen_vals.end(), 0.0);
 
@@ -213,7 +213,7 @@ void Analyze::run_offline_analysis(std::string outfile, float range, float steps
   // export modes
   std::vector<json> modes;
   for (int mode = 0; mode < num_modes; mode++) {
-    unsigned int m = stats_.Eigenvectors().cols() - (mode + 1);
+    unsigned int m = stats_.get_eigen_vectors().cols() - (mode + 1);
     json jmode;
     jmode["mode"] = mode + 1;
     double eigen_value = eigen_vals[mode];
@@ -231,8 +231,8 @@ void Analyze::run_offline_analysis(std::string outfile, float range, float steps
     jmode["cumulative_explained_variance"] = cumulative_explained_variance;
     SW_LOG("explained_variance [{}]: {:.2f}", mode, explained_variance);
     SW_LOG("cumulative_explained_variance [{}]: {:.2f}", mode, cumulative_explained_variance);
-
-    double lambda = sqrt(stats_.Eigenvalues()[m]);
+    
+    double lambda = sqrt(stats_.get_eigen_values()[m]);
 
     std::vector<json> jmodes;
     for (int i = 1; i <= half_steps; i++) {
@@ -278,7 +278,7 @@ void Analyze::run_offline_analysis(std::string outfile, float range, float steps
   }
 
   j["eigen_vectors"] = get_eigen_vectors(&stats_);
-  j["eigen_values"] = stats_.Eigenvalues();
+  j["eigen_values"] = stats_.get_eigen_values();
   j["modes"] = modes;
   j["charts"] = create_charts(&stats_);
 
@@ -322,7 +322,7 @@ Particles Analyze::get_mean_shape_points() {
     return Particles();
   }
 
-  return convert_from_combined(stats_.Mean());
+  return convert_from_combined(stats_.get_mean());
 }
 
 //---------------------------------------------------------------------------
@@ -336,30 +336,30 @@ ShapeHandle Analyze::get_mean_shape() {
 
 //---------------------------------------------------------------------------
 Particles Analyze::get_shape_points(int mode, double value) {
-  if (!compute_stats() || stats_.Eigenvectors().size() <= 1) {
+  if (!compute_stats() || stats_.get_eigen_vectors().size() <= 1) {
     return Particles();
   }
-  if (mode + 2 > stats_.Eigenvalues().size()) {
-    mode = stats_.Eigenvalues().size() - 2;
+  if (mode + 2 > stats_.get_eigen_values().size()) {
+    mode = stats_.get_eigen_values().size() - 2;
   }
-
-  unsigned int m = stats_.Eigenvectors().cols() - (mode + 1);
-
-  Eigen::VectorXd e = stats_.Eigenvectors().col(m);
-
-  double lambda = sqrt(stats_.Eigenvalues()[m]);
+  
+  unsigned int m = stats_.get_eigen_vectors().cols() - (mode + 1);
+  
+  Eigen::VectorXd e = stats_.get_eigen_vectors().col(m);
+  
+  double lambda = sqrt(stats_.get_eigen_values()[m]);
 
   std::vector<double> vals;
-  for (int i = stats_.Eigenvalues().size() - 1; i > 0; i--) {
-    vals.push_back(stats_.Eigenvalues()[i]);
+  for (int i = stats_.get_eigen_values().size() - 1; i > 0; i--) {
+    vals.push_back(stats_.get_eigen_values()[i]);
   }
   double sum = std::accumulate(vals.begin(), vals.end(), 0.0);
   double cumulation = 0;
   for (size_t i = 0; i < mode + 1; ++i) {
     cumulation += vals[i];
   }
 
-  auto temp_shape = stats_.Mean() + (e * (value * lambda));
+  auto temp_shape = stats_.get_mean() + (e * (value * lambda));
 
   return convert_from_combined(temp_shape);
 }
@@ -489,9 +489,9 @@ bool Analyze::compute_stats() {
       return false;
     }
   }
-
-  stats_.ImportPoints(points, group_ids);
-  stats_.ComputeModes();
+  
+  stats_.import_points(points, group_ids);
+  stats_.compute_modes();
 
   stats_ready_ = true;
   SW_LOG("Computed stats successfully");
@@ -522,7 +522,7 @@ Particles Analyze::convert_from_combined(const Eigen::VectorXd& points) {
 
 //---------------------------------------------------------------------------
 void Analyze::initialize_mesh_warper() {
-  int median = stats_.ComputeMedianShape(-32);  //-32 = both groups
+  int median = stats_.compute_median_shape(-32);  //-32 = both groups
 
   if (median < 0 || median >= get_shapes().size()) {
     SW_ERROR("Unable to set reference mesh, stats returned invalid median index");
@@ -555,9 +555,9 @@ void Analyze::initialize_mesh_warper() {
 int Analyze::get_num_subjects() { return shapes_.size(); }
 
 //---------------------------------------------------------------------------
-Eigen::VectorXf Analyze::get_subject_features(int subject, std::string feature_name) {
+Eigen::VectorXd Analyze::get_subject_features(int subject, std::string feature_name) {
   if (subject >= shapes_.size()) {
-    return Eigen::VectorXf();
+    return Eigen::VectorXd();
   }
 
   auto shape = shapes_[subject];
@@ -617,8 +617,8 @@ Particles Analyze::get_group_shape_particles(double ratio) {
   if (!compute_stats()) {
     return Particles();
   }
-
-  auto particles = stats_.Group1Mean() + (stats_.GroupDifference() * ratio);
+  
+  auto particles = stats_.get_group1_mean() + (stats_.get_group_difference() * ratio);
 
   return convert_from_combined(particles);
 }

Libs/Analyze/Analyze.h

@@ -57,7 +57,7 @@ class Analyze {
 
   ShapeHandle create_shape_from_points(Particles points);
 
-  Eigen::VectorXf get_subject_features(int subject, std::string feature_name);
+  Eigen::VectorXd get_subject_features(int subject, std::string feature_name);
 
   void set_group_selection(std::string group_name, std::string group1, std::string group2);
 

Libs/Analyze/Shape.cpp

@@ -572,7 +572,7 @@ void Shape::apply_feature_to_points(std::string feature, ImageType::Pointer imag
 
   int num_points = all_locals.size() / 3;
 
-  Eigen::VectorXf values(num_points);
+  Eigen::VectorXd values(num_points);
 
   int idx = 0;
   for (int i = 0; i < num_points; ++i) {
@@ -615,7 +615,7 @@ void Shape::load_feature_from_mesh(std::string feature, MeshHandle mesh) {
 
   int num_points = all_locals.size() / 3;
 
-  Eigen::VectorXf values(num_points);
+  Eigen::VectorXd values(num_points);
 
   vtkDataArray* from_array = from_mesh->GetPointData()->GetArray(feature.c_str());
   if (!from_array) {
@@ -639,10 +639,10 @@ void Shape::load_feature_from_mesh(std::string feature, MeshHandle mesh) {
 }
 
 //---------------------------------------------------------------------------
-Eigen::VectorXf Shape::get_point_features(std::string feature) {
+Eigen::VectorXd Shape::get_point_features(std::string feature) {
   auto it = point_features_.find(feature);
   if (it == point_features_.end()) {
-    return Eigen::VectorXf();
+    return Eigen::VectorXd();
   }
 
   return it->second;
@@ -683,7 +683,7 @@ std::vector<vtkSmartPointer<vtkTransform>> Shape::get_procrustes_transforms() {
 }
 
 //---------------------------------------------------------------------------
-void Shape::set_point_features(std::string feature, Eigen::VectorXf values) {
+void Shape::set_point_features(std::string feature, Eigen::VectorXd values) {
   point_features_[feature] = values;
 
   auto group = get_meshes(DisplayMode::Reconstructed);
@@ -778,7 +778,7 @@ void Shape::load_feature_from_scalar_file(std::string filename, std::string feat
     floats.push_back(line);
   }
 
-  Eigen::VectorXf values(floats.size());
+  Eigen::VectorXd values(floats.size());
   for (int i = 0; i < floats.size(); i++) {
     values[i] = floats[i];
   }

Libs/Analyze/Shape.h

@@ -158,9 +158,9 @@ class Shape {
 
   std::shared_ptr<Image> get_image_volume(std::string image_volume_name);
 
-  Eigen::VectorXf get_point_features(std::string feature);
+  Eigen::VectorXd get_point_features(std::string feature);
 
-  void set_point_features(std::string feature, Eigen::VectorXf values);
+  void set_point_features(std::string feature, Eigen::VectorXd values);
 
   void load_feature_from_scalar_file(std::string filename, std::string feature_name);
 
@@ -198,7 +198,7 @@ class Shape {
   std::vector<std::string> global_point_filenames_;
   std::vector<std::string> local_point_filenames_;
 
-  std::map<std::string, Eigen::VectorXf> point_features_;
+  std::map<std::string, Eigen::VectorXd> point_features_;
   Particles particles_;
 
   std::shared_ptr<shapeworks::Subject> subject_;

Libs/Analyze/StudioMesh.cpp

@@ -76,7 +76,7 @@ void StudioMesh::apply_feature_map(std::string name, ImageType::Pointer image) {
 
 //---------------------------------------------------------------------------
 void StudioMesh::interpolate_scalars_to_mesh(std::string name, Eigen::VectorXd positions,
-                                             Eigen::VectorXf scalar_values) {
+                                             Eigen::VectorXd scalar_values) {
   int num_points = positions.size() / 3;
   if (num_points == 0) {
     return;

Libs/Analyze/StudioMesh.h

@@ -54,7 +54,7 @@ class StudioMesh {
   void apply_scalars(MeshHandle mesh);
 
   //! Interpolation scalars at positions to this mesh
-  void interpolate_scalars_to_mesh(std::string name, Eigen::VectorXd positions, Eigen::VectorXf scalar_values);
+  void interpolate_scalars_to_mesh(std::string name, Eigen::VectorXd positions, Eigen::VectorXd scalar_values);
 
   //! Return the range of largest axis (e.g. 200 for an object that sits in 100x200x100)
   double get_largest_dimension_size();