Merge pull request #259 from thowell/num_parameters

Planner::NumParameters

Author: erez-tom

mjpc/planners/cross_entropy/planner.h

@@ -84,6 +84,11 @@ class CrossEntropyPlanner : public Planner {
   void Plots(mjvFigure* fig_planner, mjvFigure* fig_timer, int planner_shift,
              int timer_shift, int planning, int* shift) override;
 
+  // return number of parameters optimized by planner
+  int NumParameters() override {
+    return policy.num_spline_points * policy.model->nu;
+  };
+
   // ----- members ----- //
   mjModel* model;
   const Task* task;

mjpc/planners/gradient/planner.h

@@ -88,6 +88,11 @@ class GradientPlanner : public Planner {
   void Plots(mjvFigure* fig_planner, mjvFigure* fig_timer, int planner_shift,
              int timer_shift, int planning, int* shift) override;
 
+  // return number of parameters optimized by planner
+  int NumParameters() override {
+    return policy.num_spline_points * policy.model->nu;
+  };
+
   // ----- members ----- //
   mjModel* model;
   const Task* task;

mjpc/planners/ilqg/planner.cc

@@ -165,6 +165,9 @@ void iLQGPlanner::NominalTrajectory(int horizon, ThreadPool& pool) {
   if (num_trajectory_ == 0) {
     return;
   }
+  // set policy trajectory horizon
+  policy.trajectory.horizon = horizon;
+  
   // resize data for rollouts
   ResizeMjData(model, pool.NumThreads());
 

mjpc/planners/ilqg/planner.h

@@ -71,6 +71,11 @@ class iLQGPlanner : public Planner {
   void Plots(mjvFigure* fig_planner, mjvFigure* fig_timer, int planner_shift,
              int timer_shift, int planning, int* shift) override;
 
+  // return number of parameters optimized by planner
+  int NumParameters() override {
+    return policy.trajectory.dim_action * (policy.trajectory.horizon - 1);
+  };
+
   // single iLQG iteration
   void Iteration(int horizon, ThreadPool& pool);
 
@@ -85,8 +90,6 @@ class iLQGPlanner : public Planner {
 
   void UpdateNumTrajectoriesFromGUI();
 
-  //
-
   // ----- members ----- //
   mjModel* model;
   const Task* task;

mjpc/planners/ilqs/planner.h

@@ -81,6 +81,11 @@ class iLQSPlanner : public Planner {
   void Plots(mjvFigure* fig_planner, mjvFigure* fig_timer, int planner_shift,
              int timer_shift, int planning, int* shift) override;
 
+  // return number of parameters optimized by planner
+  int NumParameters() override {
+    return sampling.NumParameters() + ilqg.NumParameters();
+  };
+
   // ----- planners ----- //
   SamplingPlanner sampling;
   iLQGPlanner ilqg;

mjpc/planners/planner.h

@@ -71,6 +71,9 @@ class Planner {
                      int planner_shift, int timer_shift, int planning,
                      int* shift) = 0;
 
+  // return number of parameters optimized by planner
+  virtual int NumParameters() = 0;
+
   std::vector<UniqueMjData> data_;
   void ResizeMjData(const mjModel* model, int num_threads);
 };

mjpc/planners/robust/robust_planner.h

@@ -54,6 +54,7 @@ class RobustPlanner : public Planner {
   void GUI(mjUI& ui) override;
   void Plots(mjvFigure* fig_planner, mjvFigure* fig_timer, int planner_shift,
              int timer_shift, int planning, int* shift) override;
+  int NumParameters() override { return delegate_->NumParameters(); };
 
  private:
   void ResizeTrajectories(int ntrajectories);

mjpc/planners/sampling/planner.h

@@ -91,6 +91,11 @@ class SamplingPlanner : public RankedPlanner {
   void Plots(mjvFigure* fig_planner, mjvFigure* fig_timer, int planner_shift,
              int timer_shift, int planning, int* shift) override;
 
+  // return number of parameters optimized by planner
+  int NumParameters() override {
+    return policy.num_spline_points * policy.model->nu;
+  };
+
   // optimizes policies, but rather than picking the best, generate up to
   // ncandidates. returns number of candidates created.
   int OptimizePolicyCandidates(int ncandidates, int horizon,

mjpc/simulate.cc

@@ -526,13 +526,14 @@ void UpdateInfoText(mj::Simulate* sim,
   solerr = mju_log10(mju_max(mjMINVAL, solerr));
 
   // prepare info text
-  mju::strcpy_arr(title, "Objective\nDoFs\nControls\nTime\nMemory");
+  mju::strcpy_arr(title, "Objective\nDoFs\nControls\nParameters\nTime\nMemory");
   const mjpc::Trajectory* best_trajectory =
       sim->agent->ActivePlanner().BestTrajectory();
   if (best_trajectory) {
-    mju::sprintf_arr(content, "%.3f\n%d\n%d\n%-9.3f\n%.2g of %s",
-                     best_trajectory->total_return, m->nv, m->nu, d->time,
-                     d->maxuse_arena / (double)(d->narena),
+    int nparam = sim->agent->ActivePlanner().NumParameters();
+    mju::sprintf_arr(content, "%.3f\n%d\n%d\n%d\n%-9.3f\n%.2g of %s",
+                     best_trajectory->total_return, m->nv, m->nu, nparam,
+                     d->time, d->maxuse_arena / (double)(d->narena),
                      mju_writeNumBytes(d->narena));
   }