def test_transition_6_0(): model = LoopModel() (state0, state6) = model.get_states_by_id([0, 6]) model.set_current_state_by_state_id(6) reward = model.perform_action_a() assert model.current_state == state0 assert reward == 0
def parse_model(s):
if s == 'cm':
return ChainModel()
elif s == 'scm':
return SlipperyChainModel()
elif s == 'lm':
return LoopModel()
else:
invalid()
def test_transition_6_0():
model = LoopModel()
(state0, state6) = model.get_states_by_id([0, 6])
model.set_current_state_by_state_id(6)
reward = model.perform_action_a()
assert model.current_state == state0
assert reward == 0
def get_model(model_name):
if model_name == "Chain":
return ChainModel()
elif model_name == "SlipperyChain":
return SlipperyChainModel()
elif model_name == "Chain2":
return ChainModel2()
elif model_name == "Loop":
return LoopModel()
elif model_name == "LoopDeadEnd":
return LoopModelDeadEnd()
elif model_name == "LoopDiffTrans":
return LoopModelDiffTrans()
elif model_name == "SpecialLoop":
return SpecialLoopModel()
else:
raise Exception(model_name+ " not found")
def test_loop_a():
model = LoopModel()
(state0, state1, state2, state3,
state4) = model.get_states_by_id([0, 1, 2, 3, 4])
assert model.current_state == state0
reward = model.perform_action_a()
assert model.current_state == state1
assert reward == 0
reward = model.perform_action_a()
assert model.current_state == state2
assert reward == 0
reward = model.perform_action_a()
assert model.current_state == state3
assert reward == 0
reward = model.perform_action_a()
assert model.current_state == state4
assert reward == 0
reward = model.perform_action_a()
assert model.current_state == state0
assert reward == 1
def test_loop_b():
model = LoopModel()
(state0, state5, state6, state7,
state8) = model.get_states_by_id([0, 5, 6, 7, 8])
assert model.current_state == state0
reward = model.perform_action_b()
assert model.current_state == state5
assert reward == 0
reward = model.perform_action_b()
assert model.current_state == state6
assert reward == 0
reward = model.perform_action_b()
assert model.current_state == state7
assert reward == 0
reward = model.perform_action_b()
assert model.current_state == state8
assert reward == 0
reward = model.perform_action_a()
assert model.current_state == state0
assert reward == 2
def test_loop_b(): model = LoopModel() (state0, state5, state6, state7, state8) = model.get_states_by_id([0, 5, 6, 7, 8]) assert model.current_state == state0 reward = model.perform_action_b() assert model.current_state == state5 assert reward == 0 reward = model.perform_action_b() assert model.current_state == state6 assert reward == 0 reward = model.perform_action_b() assert model.current_state == state7 assert reward == 0 reward = model.perform_action_b() assert model.current_state == state8 assert reward == 0 reward = model.perform_action_a() assert model.current_state == state0 assert reward == 2
def test_loop_a(): model = LoopModel() (state0, state1, state2, state3, state4) = model.get_states_by_id([0, 1, 2, 3, 4]) assert model.current_state == state0 reward = model.perform_action_a() assert model.current_state == state1 assert reward == 0 reward = model.perform_action_a() assert model.current_state == state2 assert reward == 0 reward = model.perform_action_a() assert model.current_state == state3 assert reward == 0 reward = model.perform_action_a() assert model.current_state == state4 assert reward == 0 reward = model.perform_action_a() assert model.current_state == state0 assert reward == 1
from RL_game import *
from LoopModel import *
if __name__ == '__main__':
if len(sys.argv) > 1:
f = open(sys.argv[1], 'w')
run_game(LoopModel(), f)
else:
run_game(LoopModel())
