def __init__(self, nrows = 8, ncols = 8):
self.nrows = nrows
self.ncols = ncols
self.nstates = nrows * ncols
self.nactions = 4
self.left_edge = []
self.right_edge = []
self.top_edge = []
self.bottom_edge = []
self.gamma = 0.9
for x in range(self.nstates):
# note that edges are not disjoint, so we cannot use elif
if x % self.ncols == 0:
self.left_edge.append(x)
if 0 <= x < self.ncols:
self.top_edge.append(x)
if x % self.ncols == self.ncols - 1:
self.right_edge.append(x)
if (self.nrows - 1) * self.ncols <= x <= self.nstates:
self.bottom_edge.append(x)
MDP.__init__(self, nstates = self.nrows * self.ncols, nactions = 4)
def __init__(self, nrows = 8, ncols = 8):
self.nrows = nrows
self.ncols = ncols
self.nstates = nrows * ncols
self.nactions = 4
self.left_edge = []
self.right_edge = []
self.top_edge = []
self.bottom_edge = []
self.gamma = 0.9
for x in range(self.nstates):
# note that edges are not disjoint, so we cannot use elif
if x % self.ncols == 0:
self.left_edge.append(x)
if 0 <= x < self.ncols:
self.top_edge.append(x)
if x % self.ncols == self.ncols - 1:
self.right_edge.append(x)
if (self.nrows - 1) * self.ncols <= x <= self.nstates:
self.bottom_edge.append(x)
MDP.__init__(self, nstates = self.nrows * self.ncols, nactions = 4)
def __init__(self):
# actions:
# 0 == L
# 1 == R
self.gamma = 0.9
self.feature_cnt = 9
MDP.__init__(self, nstates = 4, nactions = 2)
def __init__(self, nrows = 5, ncols = 5, walls=[(1,1),(1,2),(1,3),(2,1),(2,2),(2,3),(3,1),(3,2),(3,3)], endstates = [0]):
self.nrows = nrows
self.ncols = ncols
self.walls = walls
grid = [self.coords(i) for i in range(self.nrows * self.ncols)]
grid = [s for s in grid if not s in self.walls]
self.states = dict([(i,s) for (i,s) in enumerate(grid)])
self.rstates = dict([(s,i) for (i,s) in enumerate(grid)]) # reverse lookup by grid coords
self.nstates = len(self.states)
self.nactions = 8
self.endstates = endstates
MDP.__init__(self, nstates = self.nstates, nactions = self.nactions)
def reset(self):
self.step = 0
return MDP.reset(self)