[BUG]: Evaluation logger (tensorboard callback) doesn't register mean_reward and mean_power correctly

[AlejandroCN7]

Bug 🐛

Evaluation logger (tensorboard callback) doesn't register mean_reward and mean_power correctly. This happens in LoggerEvalCallback.

To Reproduce

Thanks to @manjavacas, who saw the problem. You have to run an experiment in DRL_battery.py with tensorboard and evaluation active, and we will have the next:

The information is OK due to it is a tensorboard scale visualization issue, but mean_reward and mean_power_consumption don't appear.

The code in which it exists this error (sinergym/utils/callbacks.py):

mean_reward, std_reward = np.mean(
    episodes_rewards), np.std(episodes_rewards)
mean_ep_length, std_ep_length = np.mean(
    episodes_lengths), np.std(episodes_lengths)

self.evaluation_metrics['cumulative_reward'] = np.mean(
    mean_reward)
self.evaluation_metrics['ep_length'] = mean_ep_length
self.evaluation_metrics['power_consumption'] = np.mean(
    episodes_powers)
self.evaluation_metrics['comfort_violation(%)'] = np.mean(
    episodes_comfort_violations)
self.evaluation_metrics['comfort_penalty'] = np.mean(
    episodes_comfort_penalties)
self.evaluation_metrics['power_penalty'] = np.mean(
    episodes_power_penalties)

Expected behavior

cumulative_reward and power_consumption should be mean_reward and mean_power in tensorboard evaluation graphs.

Checklist

• I have checked that there is no similar issue in the repo (required)
• I have read the documentation (required)
• I have provided a minimal working example to reproduce the bug (required)

📝 Please, don't forget to include more labels besides bug if it is necessary.