return 1
else:
return 0
return getbetas
def finterest(env):
def getinterest(state):
if state in [0, args.len, args.len*2]:
return args.intrst
else:
return args.intrst
return getinterest
from delayed_effect import YChain
env = YChain(n=args.len)
betas = fbetas(env)
interest = finterest(env)
action_1_prob = 0.5
action_2_prob = 1-action_1_prob
fo_states = [0, args.len, 2*args.len]
v_pi = {0:0.5, args.len:2, 2*args.len:-1}
weights = np.zeros_like(np.array(env.feat).reshape(-1,1))
elif args.env == "elevator":
filename = "etd"+"_int_"+str(args.intrst)+"_env_"+str(args.env)+"_len_"+str(args.len)+"_lr_"+str(args.lr)+"_seed_"+str(args.seed)
def fbetas(env):
def getbetas(state):
if state in env.goal_states or state in env.elevator_states:
return 1