Conversation
- Introduced a minimum height for the legend HUD to ensure consistent display across game states. - Updated the bot rendering logic to include health bars above units, enhancing visual feedback on bot status. - Improved the drawing of bot figures with distinct colors for team and role, and added a helmet representation for roles. - Enhanced the rendering of unit roles and health indicators in the Pygame renderer for better clarity and user experience.
- Increased building health points and adjusted construction costs for various structures to improve gameplay balance. - Modified combat statistics for units, including health and damage values, to refine combat dynamics. - Enhanced economy settings, including resource collection rates and bot costs, to better align with game mechanics. - Updated map configurations to increase size and resource density, promoting more strategic gameplay. - Improved bot observation logic to incorporate resource capacities and enhance decision-making during gameplay. - Added new reward metrics for village performance based on food security, encouraging resource management strategies.
- Added new camera zoom parameters in default.yaml to improve user control over the 3D view. - Introduced minimum and maximum zoom levels, along with a zoom wheel step for finer adjustments. - Updated camera distance settings to enhance the overall viewing experience in the game.
…lage AI War - Introduced a new MAPPO training stage with decentralized actor and centralized critic for improved multi-agent coordination. - Added MAPPO-specific configurations and training scripts to support self-play and opponent pool management. - Enhanced bot observation logic to include both local and global state information for better decision-making. - Updated README to reflect new MAPPO features and usage instructions. - Refactored existing code to accommodate the new training environment and policies. - Added tests for MAPPO layout, environment, and policy to ensure functionality and performance.
- Clarified opponent sampling and checkpoint management in the README, specifying the need for 181-dim opponents and the organization of MAPPO training snapshots. - Adjusted training configuration to separate opponent and MAPPO-specific checkpoint directories, ensuring proper management of bot policies. - Improved logging in the MAPPO environment to provide clearer warnings about incompatible opponents and their handling during training.
- Updated the MAPPO training environment to allow all allied bots to act each tick, improving coordination and decision-making. - Introduced a new callback for tracking episode metrics, including win/loss/draw outcomes and terminal reasons, for better performance analysis. - Adjusted observation packing to support multiple local bot slots, enhancing the observation structure for the centralized critic. - Updated README and training configurations to reflect changes in observation dimensions and metrics logging. - Added tests to ensure the functionality of new features and maintain code integrity.
- Added support for human players to compete against trained MAPPO policies in a 2D environment, allowing for interactive gameplay. - Updated the README to include detailed instructions for human vs AI matches and configuration requirements. - Modified the training configuration to increase the number of initial bots for improved gameplay dynamics. - Refactored the game environment to accommodate human actions and ensure proper integration with existing bot policies. - Introduced new helper functions for collecting human actions and rendering overlays during gameplay. - Enhanced the rendering logic to display relevant information during human play, improving user experience. - Added tests to validate the new human vs AI functionality and ensure robustness of the gameplay features.
- Adjusted economy configuration to change harvest interval, amount, and food consumption values for improved gameplay balance. - Modified the EconomySystem to handle fractional food consumption correctly, ensuring accurate food deduction based on the number of alive bots. - Added a new test case to validate the correct calculation of food consumption when multiple bots are alive, enhancing test coverage for the economy system.
- Updated `pyproject.toml` to ignore specific linting errors for scripts, improving code quality checks. - Modified `README.md` to clarify gameplay instructions and training configurations, focusing on MAPPO integration. - Changed training configuration in `default.yaml` to reflect the new MAPPO training stage. - Removed outdated training configuration files to streamline the training process. - Updated `requirements.txt` by removing unnecessary dependencies, ensuring a cleaner setup. - Enhanced tensorboard plotting scripts to support new training metrics and improve visualization.
- Introduced a new script for hyperparameter tuning of MAPPO using Optuna, allowing for more efficient training configurations. - Updated `README.md` to remove outdated checkpoint information and clarify the training process for MAPPO. - Modified `train_mappo_bots.py` to return metrics from training, improving performance tracking. - Enhanced the `load_project_config` function to accept overrides, providing more flexibility in configuration management. - Cleaned up the `MAPPOBotEnv` class by removing unused opponent loading logic, streamlining the environment setup. - Updated various observation and training configurations to reflect changes in the MAPPO implementation. - Added tests to validate the new hyperparameter tuning functionality and ensure robustness of the training process.
- Updated `.gitignore` to include `dist/` and `build/` directories for cleaner builds. - Modified `pyproject.toml` to ignore specific linting errors, improving code quality checks. - Introduced CI workflows in `.github/workflows/ci.yml` for automated linting, building, and testing. - Added a release workflow in `.github/workflows/release.yml` to automate package building and asset publishing on version tags. - Refactored imports in several source files to use `collections.abc.Mapping`, improving compatibility with Python 3.10 and above.
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
No description provided.