8c17debc9dc6552da8c317c0c852b55f03c80c61,slm_lab/env/__init__.py,EnvSpace,step,#EnvSpace#Any#,258

Before Change

    def step(self, action_space):
        reward_proj = []
        state_proj = []
        done_proj = []
        for e, env in enumerate(self.envs):
            action = action_space.get(e=e)
            reward, state, done = env.step(action)
            reward_proj.append(reward)
            state_proj.append(state)
            done_proj.append(done)
        reward_space = self.aeb_space.add("reward", reward_proj)
        state_space = self.aeb_space.add("state", state_proj)
        done_space = self.aeb_space.add("done", done_proj)
        return reward_space, state_space, done_space

After Change

    def step(self, action_space):
        reward_v = self.aeb_space.data_spaces["reward"].init_data_v()
        state_v = self.aeb_space.data_spaces["state"].init_data_v()
        done_v = self.aeb_space.data_spaces["done"].init_data_v()
        for env in self.envs:
            e = env.e
            action_e = action_space.get(e=e)
            reward_e, state_e, done_e = env.step(action_e)
            reward_v[e, 0:len(reward_e)] = reward_e
            state_v[e, 0:len(state_e)] = state_e
            done_v[e, 0:len(done_e)] = done_e
        reward_space = self.aeb_space.add("reward", reward_v)
        state_space = self.aeb_space.add("state", state_v)
        done_space = self.aeb_space.add("done", done_v)

Instances