trainer.train() result: <ray.rllib.agents.trainer_template.PPO object at 0x000001FB934FAE20> -> 305 episodes
OMP: Error #15: Initializing libiomp5md.dll, but found libiomp5md.dll already initialized.
OMP: Hint This means that multiple copies of the OpenMP runtime have been linked into the program. That is dangerous, since it can degrade performance or cause incorrect results. The best thing to do is to ensure that only a single OpenMP runtime is linked into the process, e.g. by avoiding static linking of the OpenMP runtime in any library. As an unsafe, unsupported, undocumented workaround you can set the environment variable KMP_DUPLICATE_LIB_OK=TRUE to allow the program to continue to execute, but that may cause crashes or silently produce incorrect results. For more information, please see http://www.intel.com/software/products/support/.
127.0.0.1 - - [31/May/2021 14:42:30] "POST / HTTP/1.1" 200 -
Windows fatal exception: access violation
Stack (most recent call first):
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\tensorboardX\summary.py", line 250 in make_histogram
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\tensorboardX\summary.py", line 210 in histogram
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\tensorboardX\writer.py", line 546 in add_histogram
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\ray\tune\logger.py", line 217 in on_result
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\ray\tune\logger.py", line 321 in on_result
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\ray\tune\trainable.py", line 821 in _log_result
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\ray\tune\trainable.py", line 809 in log_result
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\ray\rllib\agents\trainer.py", line 601 in log_result
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\ray\tune\trainable.py", line 294 in train
File "C:\Users\Admin\anaconda3\envs\ray_tf2\lib\site-packages\ray\rllib\agents\trainer.py", line 562 in train
File "Server.py", line 129 in <module>
I tried to run the PolicyClient example from examples (cartpole_client.py and cartpole_server.py) on Windows 10 with ray 1.3.0 and TF 2.4.0. But after a few episodes I get the following error:
Pip list: