Hi everyone,
I used Tune and RLlib (release 1.13.0) and all works good, but when I include a search algorithm an error is presented:
Traceback (most recent call last):
File "c:\Users\grhen\Documents\GitHub\EP_RLlib\EPRLlib_MA-server(Tune).py", line 369, in <module>
analysis = tune.run(
File "C:\Users\grhen\AppData\Local\Programs\Python\Python39\lib\site-packages\ray\tune\tune.py", line 596, in run
if config and not searcher_set_search_properties_backwards_compatible(
File "C:\Users\grhen\AppData\Local\Programs\Python\Python39\lib\site-packages\ray\tune\suggest\util.py", line 31, in set_search_properties_backwards_compatible
File "C:\Users\grhen\AppData\Local\Programs\Python\Python39\lib\site-packages\ray\tune\suggest\bayesopt.py", line 420, in <dictcomp> bounds = {"/".join(path): resolve_value(domain) for path, domain in domain_vars}
TypeError: sequence item 1: expected str instance, int found
My configuration is:
algo = BayesOptSearch(
metric='episode_reward_mean',
mode='max',
)
analysis = tune.run(
args.run,
config=config,
stop=stop,
verbose=2,
# if you would like to collect the stream outputs in files for later analysis or
# troubleshooting, Tune offers an utility parameter, log_to_file, for this.
log_to_file=True,
# name of your experiment
name="experimento_2023-02-09_2",
# a directory where results are stored before being
# sync'd to head node/cloud storage
local_dir="C:/Users/grhen/Documents/RLforEP_Resultados",
# sync our checkpoints via rsync
# you don't have to pass an empty sync config - but we
# do it here for clarity and comparison
sync_config=sync_config,
scheduler=asha_scheduler,
search_alg=algo,
# we'll keep the best five checkpoints at all times
# checkpoints (by AUC score, reported by the trainable, descending)
checkpoint_score_attr="max-episode_reward_mean",
keep_checkpoints_num=5,
# a very useful trick! this will resume from the last run specified by
# sync_config (if one exists), otherwise it will start a new tuning run
resume= "AUTO", #True, False, to resume the experiment or not, or AUTO, which will attempt to resume the experiment if possible, and otherwise will start a new experiment.
)
There are any cloud how to fix it?
Thanks