def __init__(self, nactions, alpha, gamma, ld, nlevels = 10, resolution = 0.1):
self.nactions = nactions
self.alpha = alpha
self.gamma = gamma
self.ld = ld
self.cmac = []
self.nlevels = nlevels
self.resolution = resolution
for a in range(nactions):
self.cmac.append(TraceCMAC(self.nlevels, self.resolution, alpha, ld * gamma, replace=True, inc=1.0))
class TDCmac(TD):
def __init__(self, alpha, gamma, ld, nlevels = 10, resolution = 0.1):
self.nlevels = nlevels
self.resolution = resolution
self.cmac = TraceCMAC(self.nlevels, self.resolution, alpha, ld * gamma, replace = True, inc = 1.0)
self.gamma = gamma
def __len__(self):
return len(self.cmac)
def value(self, vector):
return self.cmac.eval(vector)
def train(self, pvector, reward, vector):
delta = self.delta(pvector, reward, vector)
self.cmac.train(pvector, delta)
return delta
def reset(self):
self.cmac.reset()
def __init__(self, alpha, gamma, ld, nlevels = 10, resolution = 0.1):
self.nlevels = nlevels
self.resolution = resolution
self.cmac = TraceCMAC(self.nlevels, self.resolution, alpha, ld * gamma, replace = True, inc = 1.0)
self.gamma = gamma
