# elsciRL Core Documentation Documentation for the core Python library that can be found at: https://github.com/pdfosborne/elsciRL ## Documentation Structure This documentation is organized into several sections, each corresponding to a major component of the elsciRL framework. The structure is as follows: - **Agents/**: Describes the different agent types (e.g., Q-Learning, DQN, Random, LLM-based) and their abstract interfaces. - **Analysis/**: Guides for analysis tools and visualizations, including variance analysis and reporting. - **Application_Suite/**: Details on the application import tool and how to add or manage RL environments. - **Encoders/**: Documentation for state, action, and object encoders, including abstract interfaces and concrete implementations. - **Environment_Setup/**: Utilities for environment and agent setup, including configuration helpers and state tracking. - **Evaluation/**: Standard reporting and evaluation modules for experiments. - **Experiments/**: Experiment orchestration, including the standard experiment class and usage examples. - **GUI/**: Guides for the elsciRL web interface and prerendering scripts. - **Instruction_Following/**: Documentation for instruction-based RL, including instruction search and optimization workflows. - **Interaction_Loops/**: Core agent-environment interaction loops and their usage. ## Notes - Each section contains both high-level overviews and detailed API documentation. - Usage examples are provided for most modules, often referencing real experiment scripts. - For advanced usage, customization, or integration, see the source code and the relevant documentation section. - This wiki is intended to support both new users and advanced developers working with elsciRL and its application suite. --- For further information, see the [GitHub repository](https://github.com/pdfosborne/elsciRL) or the main [Wiki homepage](https://github.com/pdfosborne/elsciRL-Wiki).