Bug: howto_rl_mb_001_grid_world_environment.py fails

[detlefarend]

1 Describe the bug
A clear and concise description of what the bug is.

2 To Reproduce
Steps to reproduce the behavior:

1. Run the mentioned howto...

3 Expected behavior
A clear and concise description of what you expected to happen.

4 Screenshots
If applicable, add screenshots to help explain your problem.

5 Additional Informations

5.1 Operating System

• Linux
• Windows
• Mac OS

[detlefarend]

Hi @steveyuwono, could you please take a look? Thx!!

[steveyuwono]

Hi @detlefarend , this is weird that it happens only on Ubuntu with Python 3.10. But anyway I will check this later...

[steveyuwono]

Hi @detlefarend ,

I have observed that the error occurs within PyTorch’s TorchDynamo, which seems to be attempting to use Self in a type annotation. In our Unittest, this issue arises with Python 3.10 after the release of PyTorch 2.6.0. However, when I set the PyTorch version to <2.6.0, the Unittest ran without any issues.

According to the official release notes (https://github.com/pytorch/pytorch/blob/main/RELEASE.md), there should be no compatibility issues with Python 3.10. I have also tested the setup on my PC using both Windows and Ubuntu under Python 3.10, and I did not encounter any errors. This suggests that the issue might be specific to the Python 3.10 environment in the Unittest.

What should we do here? We can either limit the PyTorch version below 2.6.0 or deactivate the related howto.

[detlefarend]

Hi @detlefarend ,

I have observed that the error occurs within PyTorch’s TorchDynamo, which seems to be attempting to use Self in a type annotation. In our Unittest, this issue arises with Python 3.10 after the release of PyTorch 2.6.0. However, when I set the PyTorch version to <2.6.0, the Unittest ran without any issues.

According to the official release notes (https://github.com/pytorch/pytorch/blob/main/RELEASE.md), there should be no compatibility issues with Python 3.10. I have also tested the setup on my PC using both Windows and Ubuntu under Python 3.10, and I did not encounter any errors. This suggests that the issue might be specific to the Python 3.10 environment in the Unittest.

What should we do here? We can either limit the PyTorch version below 2.6.0 or deactivate the related howto.

Hi @steveyuwono, let's exclude 2.6.0 from the requirements. I can take over.

[steveyuwono]

Hi @detlefarend ,
I have observed that the error occurs within PyTorch’s TorchDynamo, which seems to be attempting to use Self in a type annotation. In our Unittest, this issue arises with Python 3.10 after the release of PyTorch 2.6.0. However, when I set the PyTorch version to <2.6.0, the Unittest ran without any issues.
According to the official release notes (https://github.com/pytorch/pytorch/blob/main/RELEASE.md), there should be no compatibility issues with Python 3.10. I have also tested the setup on my PC using both Windows and Ubuntu under Python 3.10, and I did not encounter any errors. This suggests that the issue might be specific to the Python 3.10 environment in the Unittest.
What should we do here? We can either limit the PyTorch version below 2.6.0 or deactivate the related howto.

Hi @steveyuwono, let's exclude 2.6.0 from the requirements. I can take over.

Okay, thank you. Then, I will delete my branch.

[detlefarend]

Hi @detlefarend ,
I have observed that the error occurs within PyTorch’s TorchDynamo, which seems to be attempting to use Self in a type annotation. In our Unittest, this issue arises with Python 3.10 after the release of PyTorch 2.6.0. However, when I set the PyTorch version to <2.6.0, the Unittest ran without any issues.
According to the official release notes (https://github.com/pytorch/pytorch/blob/main/RELEASE.md), there should be no compatibility issues with Python 3.10. I have also tested the setup on my PC using both Windows and Ubuntu under Python 3.10, and I did not encounter any errors. This suggests that the issue might be specific to the Python 3.10 environment in the Unittest.
What should we do here? We can either limit the PyTorch version below 2.6.0 or deactivate the related howto.

Hi @steveyuwono, let's exclude 2.6.0 from the requirements. I can take over.

Okay, thank you. Then, I will delete my branch.

Sry, I didn't recognize that you already created a branch for it. I changed the requirements in branch final_changes_v2.0.0 and it perfectly solved the problem. Thank you for analyzing the problem!!