How to modify the number of game steps in each cycle？

IBM / rl-testbed-for-energyplus

Reinforcement Learning Testbed for Power Consumption Optimization using EnergyPlus

MIT License

189 stars 77 forks source link

Hi！ I can only force the environment to stop at the desired number of steps by stopping it outside the environment. for example:

for b in range(num_timesteps):
            train_num +=1
            # a = env.action_space.sample()
            action,_ = agent.take_action(state)
            print(action)
            next_state, reward, done, truncated = env.step(action)
            # state, action, reward, done
            episode_return += reward
            data_save.add(state,action,reward,done)
            # state, action, reward, next_state, done
            replay_buffer.add(state, action, reward, next_state, done)
            state = next_state
            if replay_buffer.size() > minimal_size:
                b_s, b_a, b_r, b_ns, b_d = replay_buffer.sample(batch_size)
                transition_dict = {
                        'states': b_s,
                        'actions': b_a,
                        'next_states': b_ns,
                        'rewards': b_r,
                        'dones': b_d
                        }

                agent.update(transition_dict)

i used num_timesteps to control the number of steps I want to modify the steps from the environment. What parameters should I modify?

I sincerely hope to receive help. Thanks

IBM / rl-testbed-for-energyplus

How to modify the number of game steps in each cycle？ #116