Pass in the learning model instance instead of the class

digital-pro · Dec 18, 2017 · e2bc98d · e2bc98d
1 parent c242ced
commit e2bc98d
Show file tree

Hide file tree

Showing 4 changed files with 17 additions and 47 deletions.
diff --git a/learning.py b/learning.py
@@ -67,7 +67,6 @@ def loadLearning(self, filename):
 
 class NNUltimateLearning(GenericLearning):
     STATE_TO_NUMBER_MAP = {GridStates.EMPTY: 0, GridStates.PLAYER_O: -1, GridStates.PLAYER_X: 1}
-    TABLE_LEARNING_FILE = 'table_learning.json'
 
     def __init__(self, DecisionClass=TTTBoardDecision):
         self.DecisionClass = DecisionClass
@@ -81,9 +80,9 @@ def initializeModel(self):
         self.model.add(Dense(1, activation='linear', kernel_initializer='glorot_uniform'))
         self.model.compile(loss='mean_squared_error', optimizer='adam', metrics=['accuracy'])
         plot_model(self.model, to_file='model.png')
-        #self.initialModelTraining(self.TABLE_LEARNING_FILE)
 
     def initialModelTraining(self, jsonFile):
+        # If the neural network model should be seeded from some known state/value pairs
         import os
         if os.path.isfile(jsonFile):
             self.values = json.load(open(jsonFile, 'r'))

diff --git a/plotting.py b/plotting.py
@@ -15,40 +15,10 @@ def setFontProperties():
     matplotlib.rc('text', usetex=False)
     matplotlib.rcParams.update({'font.size': 14})
 
-def visualizeStateData(stateDict, dataBind, dataKey,
-                       thresholds=None,
-                       mapTitle=None,
-                       jsonFile=None,
-                       htmlFile=None):
-    statePanda = {}
-    stateCount = 0
-    for (key, value) in stateDict.items():
-        statePanda[stateCount] = [key, value]
-        stateCount += 1
-    stateData = pd.DataFrame.from_dict(statePanda, orient='index')
-    stateData.columns = [dataKey, dataBind]
-    visualizePandasData(stateData, dataBind, dataKey, thresholds, mapTitle, jsonFile, htmlFile)
-
-def visualizePandasData(stateData,dataBind, dataKey,
-                       thresholds=None,
-                       mapTitle=None,
-                       jsonFile=None,
-                       htmlFile=None):
-    state_topo = 'https://raw.githubusercontent.com/wrobstory/vincent_map_data/master/us_states.topo.json'
-    geo_data = [{'name': 'states', 'url': state_topo, 'feature': 'us_states.geo'}]
-    vis = vincent.Map(data=stateData, geo_data=geo_data, scale=1000,
-                  projection='albersUsa', data_bind=dataBind, data_key=dataKey,
-                  map_key={'states':'properties.NAME'}, brew='RdPu')
-    vis.scales[0].type='threshold'
-    vis.scales[0].domain = thresholds
-    vis.legend(title=mapTitle)
-    if jsonFile is not None and htmlFile is not None:
-        vis.to_json(jsonFile, html_out=True, html_path=htmlFile)
-
 def drawXYPlotByFactor(dataDict, xlabel='', ylabel='', legend=None,
-                       title=None, logy=False, location=2):
+                       title=None, logy=False, location=5):
     # Assuming that the data is in the format { factor: [(x1, y1),(x2,y2),...] }
-    PLOT_STYLES = ['r^-', 'bo-', 'g^-', 'ks-', 'co-', 'ms-', 'y^-']
+    PLOT_STYLES = ['r^-', 'bo-', 'g^-', 'ks-', 'ms-', 'co-', 'y^-']
     styleCount = 0
     displayedPlots = []
     pltfn = plt.semilogy if logy else plt.plot

diff --git a/test_scripts.py b/test_scripts.py
@@ -13,20 +13,21 @@ def playTTTAndPlotResults():
     learningPlayer = RLTTTPlayer()
     randomPlayer = RandomTTTPlayer()
     results = []
-    numberOfSetsOfGames = 5
+    numberOfSetsOfGames = 50
     for i in range(numberOfSetsOfGames):
-        games = GameSequence(1000, learningPlayer, randomPlayer)
+        games = GameSequence(100, learningPlayer, randomPlayer)
         results.append(games.playGamesAndGetWinPercent())
     plotValues = {'X Win Fraction': zip(range(numberOfSetsOfGames), map(lambda x: x[0], results)),
                   'O Win Fraction': zip(range(numberOfSetsOfGames), map(lambda x: x[1], results)),
                   'Draw Fraction': zip(range(numberOfSetsOfGames), map(lambda x: x[2], results))}
-    drawXYPlotByFactor(plotValues, 'Set Number', 'Fraction')
+    drawXYPlotByFactor(plotValues, 'Number of Sets (of 100 Games)', 'Fraction', title='RL Player (X) vs. Random Player (O)')
 
 def playUltimateAndPlotResults():
-    learningPlayer = RLUTTTPlayer(NNUltimateLearning)
+    learningModel = NNUltimateLearning(UTTTBoardDecision)
+    learningPlayer = RLUTTTPlayer(learningModel)
     randomPlayer = RandomUTTTPlayer()
     results = []
-    numberOfSetsOfGames = 50
+    numberOfSetsOfGames = 60
     if os.path.isfile(LEARNING_FILE):
         learningPlayer.loadLearning(LEARNING_FILE)
     for i in range(numberOfSetsOfGames):
@@ -37,10 +38,10 @@ def playUltimateAndPlotResults():
     plotValues = {'X Win Fraction': zip(range(numberOfSetsOfGames), map(lambda x: x[0], results)),
                   'O Win Fraction': zip(range(numberOfSetsOfGames), map(lambda x: x[1], results)),
                   'Draw Fraction': zip(range(numberOfSetsOfGames), map(lambda x: x[2], results))}
-    drawXYPlotByFactor(plotValues, 'Set Number', 'Fraction')
+    drawXYPlotByFactor(plotValues, 'Number of Sets (of 100 Games)', 'Fraction',title='RL Player (O) vs. Random Player (X)')
 
 def playUltimateForTraining():
-    learningPlayer = RLUTTTPlayer(TableLearning)
+    learningPlayer = RLUTTTPlayer(TableLearning())
     randomPlayer = RandomUTTTPlayer()
     games = GameSequence(4000, learningPlayer, randomPlayer, BoardClass=UTTTBoard, BoardDecisionClass=UTTTBoardDecision)
     games.playGamesAndGetWinPercent()
@@ -60,10 +61,10 @@ def plotResultsFromFile(resultsFile):
     plotValues = {'X Win Fraction': zip(range(numberOfSetsOfGames), map(lambda x: x[0], results)),
                   'O Win Fraction': zip(range(numberOfSetsOfGames), map(lambda x: x[1], results)),
                   'Draw Fraction': zip(range(numberOfSetsOfGames), map(lambda x: x[2], results))}
-    drawXYPlotByFactor(plotValues, 'Set Number', 'Fraction')
+    drawXYPlotByFactor(plotValues, 'Number of Sets (of 100 Games)', 'Fraction', title='RL Player (O) vs. Random Player (X)')
 
 if __name__ == '__main__':
     #playTTTAndPlotResults()
     #playUltimateForTraining()
-    #playUltimateAndPlotResults()
-    plotResultsFromFile('results/ultmate_nn1_results.csv')
+    playUltimateAndPlotResults()
+    #plotResultsFromFile('results/ultimate_nn1_results_o.csv')
diff --git a/ultimateplayer.py b/ultimateplayer.py
@@ -43,8 +43,8 @@ def learnFromMove(self, prevBoardState):
         pass  # Random player does not learn from move
 
 class RLUTTTPlayer(UTTTPlayer):
-    def __init__(self, Learning=TableLearning):
-        self.learningAlgo = Learning(UTTTBoardDecision)
+    def __init__(self, learningModel):
+        self.learningAlgo = learningModel
         super(RLUTTTPlayer, self).__init__()
 
     def printValues(self):
@@ -96,4 +96,4 @@ def loadLearning(self, filename):
 if __name__  == '__main__':
     board = UTTTBoard()
     player1 = RandomUTTTPlayer()
-    player2 = RLUTTTPlayer()
+    player2 = RLUTTTPlayer(TableLearning(UTTTBoardDecision))