Closed LinBornRain closed 6 years ago
Hi, thanks for the appreciation.
I assume you follow the step here. If so, I don't think this problem should occur! Can you further elaborate your problem?
Just for reminder: please navigate to cd $GAIL-TF/baselines/ppo1
. I haven't well organized the code in $GAIL-TF/baselines/trpo_mpi
I see. maybe thats why u use ppo instead of trpo to generate policy lol. anyway ,I tried to use Pycharm to run in the file"run_mujoco.py" and it worked! however,following the step, and have cd to $GAIL-TF/baselines/ppo1 in the terminal, the terminal just fail to run the 'python run_mujoco.py' ,reporting like above.its kind of odd ,cause i can run the iteration in Pycharm. I guess i can try to modify the default number in the code instead of trying the code in terminal like 'python run_mujoco.py --env_id Ant-v1' to change the environment!
it seems that you find ur way out, lol. I'll first close the issue. Feel free to reopen it if needed.
The reason that I fail to run the code is that I have installed the openai package baselines before . Once I uninstall the package 'baselines' , the problem solved!!Thanks anyway!
[2017-10-18 19:40:13,749] I run the code in PYTHON 3.6.2 and have install the mujoco and configure the mjkey successfully:
$Making new env: Ant-v1 Traceback (most recent call last): File "run_mujoco.py", line 56, in
main()
File "run_mujoco.py", line 52, in main
train(args)
File "run_mujoco.py", line 34, in train
task_name=task_name
TypeError: learn() got an unexpected keyword argument 'ckpt_dir'
can u help me with this issue? Appreciate it that u share GAIL code in tf BTW!