4342569bc5e505a9d1a36b3b4a3d1269d3a2bcab,slm_lab/agent/algorithm/hydra_dqn.py,MultitaskDQN,space_act,#MultitaskDQN#Any#,46

Before Change

            if self.normalize_state:
                state = policy_util.update_online_stats_and_normalize_state(body, state)
            states.append(state)
        state = torch.tensor(states, device=self.net.device).view(-1).unsqueeze_(0).float()
        pdparam = self.calc_pdparam(state, evaluate=False)
        // use multi-policy. note arg change
        action_a, action_pd_a = self.action_policy(states, self, self.agent.nanflat_body_a, pdparam)

After Change

            if self.normalize_state:
                state = policy_util.update_online_stats_and_normalize_state(body, state)
            states.append(state)
        xs = [torch.from_numpy(state).float() for state in states]
        pdparam = self.calc_pdparam(xs, evaluate=False)
        // use multi-policy. note arg change
        action_a, action_pd_a = self.action_policy(states, self, self.agent.nanflat_body_a, pdparam)

Instances