About Using Unity:ML-Agents and DQN Algorithm

Viewed 91

I am having difficulty learning by connecting the external API and the unity environment I have created.

I was looking at the previous ml-agent version of DQN code and wanted to use the following code. How should I use this in the current version?

# send the action to the environment and receive resultant environment information
        env_info = env.step(action)[brain_name]        

        next_state = env_info.vector_observations[0]   # get the next state
        reward = env_info.rewards[0]                   # get the reward
        done = env_info.local_done[0]                  # see if episode has finished

and

# reset the unity environment at the beginning of each episode
    env_info = env.reset(train_mode=True)[brain_name]     

    # get initial state of the unity environment 
    state = env_info.vector_observations[0]

Like this.

I am trying to change it to the current version while watching the official documentary, but it is not clearly resolved.

How should I use this in the current ml-agents version? And what exactly does this mean? I mean, why is that becoming a 'state' of Environment?

The code below was written after watching the official documentary, but I don't know if this is the right way and I don't know the meaning.

    decision_steps, terminal_steps = env.get_steps(behavior_name)
    
    state = decision_steps.obs[index][0,:]
0 Answers
Related