print("-----------")
if save_gif:
if verbose:
print("Saving the gif of the episodes")
save_log_gif(logs_path, res)
return agent, res
if __name__ == "__main__":
from grid2op.Reward import L2RPNSandBoxScore, L2RPNReward
from l2rpn_baselines.utils import cli_eval
# Parse command line
args = cli_eval().parse_args()
# Create dataset env
env = make(args.env_name,
reward_class=L2RPNSandBoxScore,
other_rewards={"reward": L2RPNReward})
# Call evaluation interface
evaluate(env,
name=args.name,
load_path=os.path.abspath(args.load_path),
logs_path=args.logs_dir,
nb_episode=args.nb_episode,
nb_process=args.nb_process,
max_steps=args.max_steps,
verbose=args.verbose,
nb_process=nb_process,
max_iter=max_steps,
pbar=False)
# Print summary
print("Evaluation summary:")
for _, chron_name, cum_reward, nb_time_step, max_ts in res:
msg_tmp = "\tFor chronics located at {}\n".format(chron_name)
msg_tmp += "\t\t - cumulative reward: {:.6f}\n".format(cum_reward)
msg_tmp += "\t\t - number of time steps completed: {:.0f} / {:.0f}".format(
nb_time_step, max_ts)
print(msg_tmp)
if save_gif:
save_log_gif(load_path, res)
if __name__ == "__main__":
import grid2op
from l2rpn_baselines.utils import cli_eval
args_cli = cli_eval().parse_args()
env = grid2op.make()
evaluate(env,
load_path=args_cli.load_path,
logs_path=args_cli.logs_path,
nb_episode=args_cli.nb_episode,
nb_process=args_cli.nb_process,
max_steps=args_cli.max_steps,
verbose=args_cli.verbose,
save_gif=args_cli.save_gif)
