Added new methods for Skewb and Pyraminx

README.md

@@ -35,7 +35,7 @@ Currently available environments:
 | Actions | L, R, U, B |  
 | Rewards | (-inf, 100] |
 | Max steps | 250 |
-| Reward Types | Base |
+| Reward Types | Base, Layer by Layer Method |
 | Render Modes | 'human', 'rgb_array', 'ansi' |
 
 ### Skewb
@@ -47,7 +47,7 @@ Currently available environments:
 | Actions | L, R, U, B |  
 | Rewards | (-inf, 100] |
 | Max steps | 250 |
-| Reward Types | Base |
+| Reward Types | Base, Sarah's Method(Advanced) |
 | Render Modes | 'human', 'rgb_array', 'ansi' |
 
 ## Installation

rubiks_cube_gym/__init__.py

@@ -24,8 +24,20 @@
     max_episode_steps=250,
 )
 
+register(
+    id='pyraminx-wo-tips-lbl-v0',
+    entry_point='rubiks_cube_gym.envs:PyraminxWoTipsEnvLBL',
+    max_episode_steps=250,
+)
+
 register(
     id='skewb-v0',
     entry_point='rubiks_cube_gym.envs:SkewbEnv',
     max_episode_steps=250,
+)
+
+register(
+    id='skewb-sarah-v0',
+    entry_point='rubiks_cube_gym.envs:SkewbEnvSarah',
+    max_episode_steps=250,
 )

rubiks_cube_gym/envs/__init__.py

@@ -2,4 +2,6 @@
 from rubiks_cube_gym.envs.rubiks_cube_222_lbl import RubiksCube222EnvLBL
 from rubiks_cube_gym.envs.rubiks_cube_222_ortega import RubiksCube222EnvOrtega
 from rubiks_cube_gym.envs.pyraminx_wo_tips import PyraminxWoTipsEnv
+from rubiks_cube_gym.envs.pyraminx_wo_tips_lbl import PyraminxWoTipsEnvLBL
 from rubiks_cube_gym.envs.skewb import SkewbEnv
+from rubiks_cube_gym.envs.skewb_sarah import SkewbEnvSarah

rubiks_cube_gym/envs/pyraminx_wo_tips_lbl.py

@@ -0,0 +1,51 @@
+from rubiks_cube_gym.envs.pyraminx_wo_tips import PyraminxWoTipsEnv
+import numpy as np
+from operator import itemgetter
+
+
+class PyraminxWoTipsEnvLBL(PyraminxWoTipsEnv):
+    def __init__(self):
+        super(PyraminxWoTipsEnvLBL, self).__init__()
+        self.FL = None
+        self.OLL = None
+
+    def check_FL(self):
+        for pos in FL_POS:
+            if itemgetter(*pos)(self.cube_reduced) == itemgetter(*pos)("RRRRRGBBBBBRRRGGGBBBRGGGGGBYYYYYYYYY"):
+                return True
+        return False
+
+    def check_solved(self):
+        if self.cube_reduced == "RRRRRGBBBBBRRRGGGBBBRGGGGGBYYYYYYYYY":
+            return True
+
+    def reward(self):
+        reward = -40 * self.FL
+        done = False
+
+        if self.check_FL():
+            reward += 40
+            self.FL = True
+        else:
+            self.FL = False
+
+        if self.check_solved():
+            reward += 60
+            done = True
+
+        if reward <= 0:
+            reward -= 1
+
+        return reward, done
+
+    def reset(self, scramble=None):
+        super(PyraminxWoTipsEnvLBL, self).reset(scramble=scramble)
+        self.FL = self.check_FL()
+
+        return self.cube_state
+
+
+FL_POS = [[0, 1, 2, 3, 4, 11, 12, 13, 20, 5, 14, 15, 20, 21, 6, 7, 8, 9, 10, 27, 28, 32, 33, 35],
+          [5, 14, 15, 16, 21, 22, 23, 24, 25, 3, 4, 12, 13, 20, 6, 7, 17, 18, 26, 27, 28, 29, 30, 31],
+          [6, 7, 8, 9, 10, 17, 18, 19, 26, 5, 15, 16, 24, 25, 30, 31, 33, 34, 35, 0, 1, 2, 3, 4],
+          [27, 28, 29, 30, 31, 32, 33, 34, 35, 21, 22, 23, 24, 25, 0, 1, 11, 12, 20, 9, 10, 18, 19, 26]]

rubiks_cube_gym/envs/skewb_sarah.py

@@ -0,0 +1,49 @@
+from rubiks_cube_gym.envs.skewb import SkewbEnv
+import numpy as np
+from operator import itemgetter
+
+
+class SkewbEnvSarah(SkewbEnv):
+    def __init__(self):
+        super(SkewbEnvSarah, self).__init__()
+        self.FL = None
+
+    def check_FL(self):
+        for pos in FL_POS:
+            if itemgetter(*pos)(self.cube_reduced) == itemgetter(*pos)("WWWWWOOOOOGGGGGRRRRRBBBBBYYYYY"):
+                return True
+        return False
+
+    def check_solved(self):
+        if self.cube_reduced == "WWWWWOOOOOGGGGGRRRRRBBBBBYYYYY":
+            return True
+
+    def reward(self):
+        reward = -40 * self.FL
+        done = False
+
+        if self.check_FL():
+            reward += 40
+            self.FL = True
+        else:
+            self.FL = False
+
+        if self.check_solved():
+            reward += 60
+            done = True
+
+        if reward <= 0:
+            reward -= 1
+
+        return reward, done
+
+    def reset(self, scramble=None):
+        super(SkewbEnvSarah, self).reset(scramble=scramble)
+        self.FL = self.check_FL()
+
+        return self.cube_state
+
+
+FL_POS = [[0, 1, 2, 3, 4, 5, 6, 10, 11, 15, 16, 20, 21], [5, 6, 7, 8, 9, 0, 3, 10, 13, 25, 28, 21, 24],
+          [10, 11, 12, 13, 14, 3, 4, 15, 18, 25, 26, 6, 9], [15, 16, 17, 18, 19, 1, 4, 20, 23, 26, 29, 11, 14],
+          [20, 21, 22, 23, 24, 16, 19, 28, 29, 5, 8, 0, 1], [25, 26, 27, 28, 29, 8, 9, 13, 14, 18, 19, 23, 24]]

setup.py

@@ -4,7 +4,7 @@
     long_description = f.read()
 
 setup(name='rubiks_cube_gym',
-      version='0.2.1',
+      version='0.3.1',
       url="https://github.com/DoubleGremlin181/RubiksCubeGym/",
       description="OpenAI Gym environments for various twisty puzzles",
       long_description=long_description,