rical730
rical730
Error occur when using xparl and setting cpu_num too large. Such as: `xparl start --port 8344 --monitor_port 8345 --cpu_num 100` E0111 17:22:04.162848529 1125 ev_epollex_linux.cc:1302] pollset_add_fd: {"created":"@1641892924.162793536","description":"pollset_transition_pollable_from_empty_to_fd","file":"src/core/lib/iomgr/ev_epollex_linux.cc","file_line":320,"referenced_errors":[{"created":"@1641892924.162791751","description":"get_fd_pollable","file":"src/core/lib/iomgr/ev_epollex_linux.cc","file_line":320,"referenced_errors":[{"created":"@1641892924.162783652","description":"Too many open files","errno":24,"file":"src/core/lib/iomgr/wakeup_fd_eventfd.cc","file_line":38,"os_error":"Too many...
Is there a more convenient way for parameter tuning and configuration?
 Can we remind users more friendly when the remote job ends normally?
When loading parameters, paddle2.0 will skip the parameters of the wrong shape and just throw a warning. PARL will add patchs to prevent this kind of situation and report an...
Hi, I can successfully run qmix experiment ``` bash run.sh $GPU python3 src/main.py --config=qmix --env-config=sc2 with env_args.map_name=2s3z ``` but errors occurred when I replace qmix with coma ``` bash run.sh...
Python (3.5.4) OpenAI gym (0.10.5) tensorflow (1.8.0) numpy (1.14.5) maddpg code: https://github.com/openai/maddpg I got an error when I run the maddpg code in the simple scene and display it PS:...
Can we output the error log of the remote actor to the monitoring web page？It's not convenient to debug when I am using xparl in jupyter
I have completed the entire continuous integration process based on Github Actions. Remaining issues have been marked as TODO in `.teamcity/action_build.sh`. Please take note. More details in https://github.com/PaddlePaddle/PARL/pull/1139