def __init__(self, alphaStepSize=1e-4, **kwargs):
SARSA.__init__(self, alphaStepSize=1e-4, **kwargs)
self.numFeatures = self.numRays + 1
self.initializeWeights()
self.resetElibilityTraces()
grid = np.arange(0, self.numRays)
self.plotGrid = grid - np.mean(grid)
def __init__(self, alphaStepSize=1e-4, **kwargs):
SARSA.__init__(self, alphaStepSize=1e-4, **kwargs)
self.numFeatures = self.numRays + 1
self.initializeWeights()
self.resetElibilityTraces()
grid = np.arange(0, self.numRays)
self.plotGrid = grid - np.mean(grid)
def __init__(self, numInnerBins=4, numOuterBins=4, binCutoff=0.5, alphaStepSize=0.2,
useQLearningUpdate= False, **kwargs):
SARSA.__init__(self, alphaStepSize=0.2, **kwargs)
self.numInnerBins=numInnerBins
self.numOuterBins=numOuterBins
self.numBins=numInnerBins + numOuterBins
self.binCutoff=binCutoff
self.useQLearningUpdate = useQLearningUpdate
self.initializeQValues()
self.initializeBinData()
self.resetElibilityTraces()
self.eligibilityTraceThreshold = 0.1
def __init__(self, numInnerBins=4, numOuterBins=4, binCutoff=0.5, alphaStepSize=0.2, forceDriveStraight=False,
useQLearningUpdate= False, **kwargs):
SARSA.__init__(self, alphaStepSize=0.2, **kwargs)
self.numInnerBins=numInnerBins
self.numOuterBins=numOuterBins
self.numBins=numInnerBins + numOuterBins
self.binCutoff=binCutoff
self.useQLearningUpdate = useQLearningUpdate
self.forceDriveStraight = forceDriveStraight
self.initializeQValues()
self.initializeBinData()
self.resetElibilityTraces()
self.eligibilityTraceThreshold = 0.1
