Deepmind hard solved

Yao-Chung · Nov 13, 2021 · afbc75b · afbc75b
1 parent 28b824b
commit afbc75b
Show file tree

Hide file tree

Showing 3 changed files with 86 additions and 21 deletions.
diff --git a/exec/agent1.cpp b/exec/agent1.cpp
@@ -28,23 +28,85 @@ int main(int argc, char* const argv[])
     // }
 
     // FIXME: Hard-coded map
-    std::vector<std::string> level = {
-        "########",
-        "#. #   #",
-        "#  $   #",
-        "#   # ##",
-        "## # $.#",
-        "#   $  #",
-        "#  .# @#",
-        "########",
+    std::vector<std::vector<std::string>> levels = {
+        {
+            // Teacher
+            "########",
+            "#. #   #",
+            "#  $   #",
+            "#   # ##",
+            "## # $.#",
+            "#   $  #",
+            "#  .# @#",
+            "########",
+        },
+        {
+            // Medium
+            "##########",
+            "# ###### #",
+            "# #####. #",
+            "# #####  #",
+            "#  ### $ #",
+            "#@$    . #",
+            "# .# $.  #",
+            "# $####  #",
+            "#    #####",
+            "##########",
+        },
+        {
+            // Medium
+            "##########",
+            "# ###    #",
+            "# @## $. #",
+            "# $. .$$ #",
+            "#   # .  #",
+            "##########",
+        },
+        {
+            // Medium
+            "##########",
+            "###@######",
+            "# .$### ##",
+            "# $  . $ #",
+            "#    . $ #",
+            "## ## #  #",
+            "#####  # #",
+            "#### .#  #",
+            "####     #",
+            "##########",
+        },
+        {
+            // Hard
+            "#######",
+            "##### #",
+            "#@ ## #",
+            "# $   #",
+            "## $  #",
+            "#  #..#",
+            "# $$  #",
+            "# .  .#",
+            "#######",
+        },
+        {
+            // Hard
+            "##########",
+            "######   #",
+            "#######$ #",
+            "####### .#",
+            "#  ### $ #",
+            "#.$ #    #",
+            "#  @#  ###",
+            "# $ ##  ##",
+            "#  ..    #",
+            "##########",
+        },
     };
+    std::vector<std::string> &level = levels[5];
 
     // Create solver
     Solver solver(level, prefix, extension);
     // Solve
     std::vector<MoveDirection> policy = solver.solve();
-    // Optimize policy
-    policy = PolicyOptimize::optimize(level, policy);
     // Replay policy
     std::cout << "== Replay ==" << std::endl;
     std::vector<std::string> map(level);

diff --git a/lib/CMakeLists.txt b/lib/CMakeLists.txt
@@ -3,6 +3,5 @@ add_library(sokoban-agent
     Solver.cpp
     defines.cpp
     Visualizer.cpp
-    PolicyOptimize.cpp
     State.cpp
 )
diff --git a/lib/Solver.cpp b/lib/Solver.cpp
@@ -34,6 +34,14 @@ static inline std::string getBoxKey(const std::string& key){
     return key.substr(key.find("B:"));
 }
 
+static void printMap(const Map &map){
+    // Print map
+    for(std::string row: map){
+        std::cout << row << std::endl;
+    }
+    std::cout << std::endl;
+}
+
 std::vector<MoveDirection> Solver::solve(){
     clean();
     // Initialize current map and current state
@@ -44,6 +52,7 @@ std::vector<MoveDirection> Solver::solve(){
     // Visualize
     visualize(0, curState, map);
     for(unsigned int iteration=1; ; ++iteration){
+        // Check fresh state
         if(curState->childs.empty()){
             // Make actions
             for(MoveDirection dir: {
@@ -102,8 +111,6 @@ std::vector<MoveDirection> Solver::solve(){
             }
             // Check if curState is dead node
             if(curState->childs.empty()){
-                // Visualize
-                visualize(0, curState, map);
                 // Clean dead nodes
                 while (curState->childs.empty()){
                     if(curState->parent == nullptr){
@@ -140,14 +147,15 @@ std::vector<MoveDirection> Solver::solve(){
         // Move to the next state
         curState = nextState;
         map = move(map, nextDir, level);
-        // Visualize
-        visualize(0, curState, map);
         // Check if reach the maxIter
         if(iteration >= maxIter){
             // Update alpha, beta, maxIter
             alpha = (Decimal)maxIter / (Decimal)(restartCount + 1);
             beta = ((curState->finishTargets == 0) ? 1 : ((Decimal)maxIter / (Decimal)curState->finishTargets)) + (Decimal)boxMoveCount / (Decimal)maxIter;
             maxIter += deltaIter;
+            std::cerr << "MaxIter: " << maxIter << std::endl; 
+            // Visualize
+            visualize(0, curState, map);
             curState = restart(map, iteration, curState);
         }
     }
@@ -217,18 +225,14 @@ MoveDirection Solver::decide(const State* const state){
 
 void Solver::visualize(const unsigned int iteration, const State* const curState, const Map& map){
     if(visualizer.has_value()){
-        // Print map
-        for(std::string row: map){
-            std::cout << row << std::endl;
-        }
         // Copy policy to set
         std::unordered_set<const State*> policy;
         for(const State* cursor = curState; cursor != nullptr; cursor = cursor->parent){
             policy.insert(cursor);
         }
 
         // Prologue
-        visualizer->next(0);
+        visualizer->next(1);
         visualizer->out << std::string("digraph{") << std::endl;
         // Print states by BFS
         std::queue<State*> stateQueue;