def feed(self, ts):
state, action, reward, next_state, done = tuple(ts)
state = wrap_state(state)
next_state = wrap_state(next_state)
ts = (state, action, reward, next_state, done)
return super().feed(ts)
def step(self, state):
return super().step(wrap_state(state))
def eval_step(self, state):
return super().eval_step(wrap_state(state))