def policy(self, float_state_features, int_state_features, actions):
float_examples = []
for i in range(len(float_state_features)):
float_examples.append({**float_state_features[i], **actions[i]})
if int_state_features is None:
return RLPredictor.policy(self, float_examples)
return RLPredictor.policy(self, float_examples, int_state_features)
def policy(self, states, actions):
examples = []
for i in range(len(states)):
examples.append({**states[i], **actions[i]})
return RLPredictor.policy(self, examples)