-
Notifications
You must be signed in to change notification settings - Fork 1.7k
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
[Bug]: Load Trained Policy #1911
Comments
Hello, Could you share a google colab where you can reproduce the error? |
Same to me. |
Hello! @araffin. sb3.get_system_info()
Freezeabsl-py==1.4.0 AdditionFrom #1852 , I downgraded a Stable-baselines3 version from 2.3.0 to 2.0.0, to match a torch version from 1.13.1 to 1.12.1. Then the warning disappeared with working well |
@CAI23sbP it seems you are also using outdated packages, could you please upgrade/remove the one i mentioned in my comment and check each time if it solves the issue? (gym should be removed, torch, cloudpickle upgraded) if it doesn't work, please start from a fresh virtual env. |
@kit1980 I could reproduce the issue, it seems to be linked to using an old version of pytorch (1.13.1). If you upgrade to PyTorch >= 2.0, problem is solved. |
Maybe Python version is the problem, not PyTorch? |
OK, per reply there, 71 is BINFLOAT op, which as added to |
🐛 Bug
Even the example code in Doc runs into bugs, loading is the problem
To Reproduce
Relevant log output / Error message
System Info
Checklist
The text was updated successfully, but these errors were encountered: