Skip to content
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

Remove or Adjust Unnecessary Warning Messages for Default goal_region Pose #708

Open
songyuc opened this issue Nov 20, 2024 · 0 comments
Open

Comments

@songyuc
Copy link

songyuc commented Nov 20, 2024

Hi,
When running the PPO examples in ManiSkill, particularly the ppo_rgb.py script, I have noticed that the following warning message appears frequently during execution:

2024-11-20 21:30:17,567 - mani_skill  - WARNING - No initial pose set for actor builder of goal_region, setting to default pose q=[1,0,0,0], p=[0,0,0].
2024-11-20 21:30:20,167 - mani_skill  - WARNING - No initial pose set for actor builder of goal_region, setting to default pose q=[1,0,0,0], p=[0,0,0].

Problem Description:

  1. Redundant Information:

    • The warning indicates that the goal_region does not have an initial pose set and is therefore using default values (q=[1,0,0,0], p=[0,0,0]). However, in many cases, the default pose is sufficient for running the examples and training the models.
  2. Cluttered Logs:

    • These warnings appear repeatedly, which can clutter the log output and make it harder to spot more critical messages or errors. This can be particularly problematic when users are trying to debug or monitor the training process.
  3. Default Pose Sufficiency:

    • For most users, especially those running the examples for the first time or during development and testing, the default pose is adequate. The warnings do not provide additional actionable information and can be seen as noise.

Proposed Solutions:

  1. Remove the Warnings:

    • If the default goal_region pose is sufficient for most use cases, consider removing these warnings entirely. This would streamline the log output and reduce unnecessary verbosity.
  2. Adjust Warning Level:

    • Alternatively, if the information is still deemed useful for some users, consider changing the warning level to INFO or DEBUG. This way, users who wish to see these messages can adjust their logging configuration accordingly, while others can avoid being overwhelmed by non-critical information.
    # Example of changing the log level
    logging.getLogger('mani_skill').setLevel(logging.INFO)  # or logging.DEBUG
  3. Provide Configuration Option:

    • Introduce a configuration option that allows users to enable or disable these warnings. This can be done through a configuration file or a command-line argument. For example:

      # config.yaml
      warn_default_pose: False  # Set to True to enable warnings
      # Command-line argument example
      python ppo_rgb.py --warn_default_pose=False
  4. Documentation Update:

    • Update the documentation to explain the default goal_region pose and inform users that they can adjust it if needed. This would help users understand why the warnings appear and how to configure them.

Example of Proposed Code Change:

# Current warning
logging.warning("No initial pose set for actor builder of goal_region, setting to default pose q=[1,0,0,0], p=[0,0,0].")

# Proposed change: Adjust warning level or remove it
# Option 1: Change to INFO level
logging.info("No initial pose set for actor builder of goal_region, setting to default pose q=[1,0,0,0], p=[0,0,0].")

# Option 2: Remove the warning entirely
# No code needed, just ensure the warning is not triggered

Conclusion:

Removing or adjusting these warnings would improve the user experience by making the logs cleaner and more focused on important information. It would also reduce confusion for new users who might not understand the relevance of the default goal_region pose. This change would make ManiSkill more user-friendly and the logs more informative.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

No branches or pull requests

1 participant