

In [64]:

    
from baselines.common import plot_util as pu

If you want to average results for multiple seeds, LOG_DIRS must contain subfolders in the following format: <name_exp0>-0, <name_exp0>-1, <name_exp1>-0, <name_exp1>-1. Where names correspond to experiments you want to compare separated with random seeds by dash.



In [80]:

    
LOG_DIRS = 'logs/reacher/'
# Uncomment below to see the effect of the timit limits flag
# LOG_DIRS = 'time_limit_logs/reacher'



In [81]:

    
results = pu.load_results(LOG_DIRS)









    



/home/kostrikov/GitHub/baselines/baselines/bench/monitor.py:163: UserWarning: Pandas doesn't allow columns to be created via a new attribute name - see https://pandas.pydata.org/pandas-docs/stable/indexing.html#attribute-access
  df.headers = headers # HACK to preserve backwards compatibility



In [82]:

    
fig = pu.plot_results(results, average_group=True, split_fn=lambda _: '', shaded_std=False)



In [ ]: