GeneralizedNotationNotation (GNN)

A standardized text-based language for Active Inference generative models

🚀 Quick Start • 📖 Documentation • 🎯 Examples • 🛠️ Tools • 🤝 Contributing

📋 Table of Contents

🌟 Overview
🎯 Motivation and Goals
✨ Key Features
🏗️ Project Architecture
⚙️ Processing Pipeline
🛠️ Tools and Utilities
🚀 Quick Start
📖 Documentation
🎯 Examples
🤝 Contributing
📄 License

🌟 Overview

Generalized Notation Notation (GNN) is a text-based language designed to standardize the representation and communication of Active Inference generative models. It aims to enhance clarity, reproducibility, and interoperability in the field of Active Inference and cognitive modeling.

📚 Initial Publication

Smékal, J., & Friedman, D. A. (2023). Generalized Notation Notation for Active Inference Models. Active Inference Journal.
📖 DOI: 10.5281/zenodo.7803328
📁 Archive: zenodo.org/records/7803328

🎯 Core Design Principles

GNN provides a structured and standardized way to describe complex cognitive models. It is designed to be:

🧑‍💻 Human-readable: Easy to understand and use for researchers from diverse backgrounds
🤖 Machine-parsable: Can be processed by software tools for analysis, visualization, and code generation
🔄 Interoperable: Facilitates the exchange and reuse of models across different platforms and research groups
🔬 Reproducible: Enables precise replication of model specifications

GNN addresses the challenge of communicating Active Inference models, which are often described using a mix of natural language, mathematical equations, diagrams, and code. By offering a unified notation, GNN aims to streamline collaboration, improve model understanding, and accelerate research.

🎯 Motivation and Goals

🚫 Current Challenges

The primary motivation behind GNN is to overcome the limitations arising from the lack of a standardized notation for Active Inference models. This fragmentation can lead to difficulties in:

💬 Effective Communication: Making complex models hard to explain and understand
🔄 Reproducibility: Hindering the ability to replicate research findings
⚙️ Consistent Implementation: Leading to variations when translating models into code
📊 Systematic Comparison: Making it challenging to compare different models

🎯 Our Goals

The goals of GNN are to:

✅ Facilitate clear communication and understanding of Active Inference models
✅ Promote collaboration among researchers
✅ Enable the development of tools for model validation, visualization, and automated code generation
✅ Support the creation of a shared repository of Active Inference models
✅ Bridge the gap between theoretical concepts and practical implementations

✨ Key Features

🎭 The Triple Play Approach

GNN supports three complementary modalities for model representation, known as the "Triple Play":

graph LR
    subgraph "🎭 Triple Play Approach"
        A["📝 Text-Based Models<br/>• GNN Markdown files<br/>• Human-readable syntax<br/>• Mathematical notation<br/>• Natural language descriptions"]
        
        B["📊 Graphical Models<br/>• Factor graphs<br/>• Network visualizations<br/>• Dependency diagrams<br/>• Interactive visualizations"]
        
        C["⚙️ Executable Models<br/>• PyMDP simulations<br/>• RxInfer.jl implementations<br/>• JAX computations<br/>• DisCoPy diagrams"]
    end
    
    A -->|Parse & Extract| B
    A -->|Generate Code| C
    B -->|Validate Structure| C
    C -->|Simulate & Test| A
    B -->|Visual Feedback| A
    C -->|Results Analysis| B
    
    style A fill:#e8f5e8,stroke:#4caf50
    style B fill:#e3f2fd,stroke:#2196f3
    style C fill:#fff3e0,stroke:#ff9800

📝 Text-Based Models: GNN files are plain text and can be rendered into mathematical notation, pseudocode, or natural language descriptions. This forms the core representation.
📊 Graphical Models: The structure defined in GNN (variables and their connections) can be visualized as graphical models (e.g., factor graphs), clarifying dependencies and model architecture.
⚙️ Executable Cognitive Models: GNN specifications can serve as a high-level blueprint or pseudocode for implementing executable simulations in various programming environments. This ensures consistency and aids in the translation from theory to practice.

📋 Structured File Format

GNN defines a specific file structure, typically using Markdown, to organize model components. This includes sections for:

🏷️ Model metadata (name, version, annotations)
🌐 State space (variable definitions)
🔗 Connections (relationships between variables)
⚙️ Initial parameterization
📐 Equations
⏰ Time settings (for dynamic models)
🧠 Mapping to Active Inference Ontology terms

🏗️ Project Architecture

graph TB
    subgraph "🏗️ GNN Project Architecture"
        subgraph "📁 Source Code (src/)"
            A[⚙️ Pipeline Scripts<br/>0_template.py → 22_mcp.py]
            B[🧠 Core Modules<br/>gnn/, render/, llm/]
            C[🔧 Utilities<br/>utils/, pipeline/]
            D[🧪 Testing<br/>tests/]
        end
        
        subgraph "📚 Documentation (doc/)"
            E[📖 Core Docs<br/>gnn/, syntax, examples]
            F[🎯 Specialized<br/>pymdp/, rxinfer/, mcp/]
            G[🧩 Applications<br/>cognitive_phenomena/]
        end
        
        subgraph "🎯 Outputs (output/)"
            H[📊 Reports<br/>Type checking, analysis]
            I[🎨 Visualizations<br/>Graphs, matrices]
            J[💻 Generated Code<br/>PyMDP, RxInfer]
            K[🌐 Static Site<br/>HTML summaries]
        end
    end
    
    A --> H
    B --> I
    B --> J
    E --> A
    F --> B
    G --> B
    
    style A fill:#e3f2fd
    style B fill:#f3e5f5
    style E fill:#e8f5e8
    style H fill:#fff3e0

📁 Directory Structure

📂 src/ Directory Structure

The src/ directory contains the 23-step pipeline scripts (0_template.py → 22_mcp.py), their corresponding modules, and shared infrastructure. See DOCS.md and doc/pipeline/README.md for the full step-by-step mapping.

src/
├── 📜 Pipeline Scripts (0-22)
│   ├── 0_template.py … 22_mcp.py   # Thin orchestrators (0–22)
├── 🧠 Core Modules
│   ├── gnn/ render/ execute/ llm/ visualization/ export/ type_checker/ ontology/ mcp/
│   ├── setup/ tests/ website/ audio/ analysis/ integration/ security/ research/ report/
├── 🔧 Infrastructure: utils/ pipeline/
└── 🗂️ Orchestrator: main.py

📂 doc/ Directory Structure

The doc/ directory contains all supplementary documentation, including conceptual explanations, syntax guides, and examples.

doc/
├── 📖 Core Documentation
│   ├── gnn/                       # GNN specifications
│   ├── quickstart.md             # Getting started guide
│   ├── SETUP.md                  # Installation instructions
│   └── README.md                 # Documentation overview
├── 🎯 Domain Applications
│   ├── cognitive_phenomena/       # Cognitive modeling examples
│   ├── pymdp/                    # PyMDP integration
│   ├── rxinfer/                  # RxInfer.jl integration
│   └── templates/                # Model templates
├── 🛠️ Technical Integration
│   ├── mcp/                      # Model Context Protocol
│   ├── llm/                      # LLM integration
│   ├── discopy/                  # DisCoPy categorical diagrams
│   └── sympy/                    # SymPy mathematical processing
└── 📚 Resources
    ├── troubleshooting/          # Common issues & solutions
    ├── testing/                  # Testing documentation
    └── security/                 # Security guidelines

⚙️ Processing Pipeline

The GNN processing pipeline consists of 23 comprehensive steps (0-22), each handling a specific aspect of model processing from parsing to final report generation. The pipeline follows a thin orchestrator pattern where numbered scripts orchestrate execution while delegating core functionality to modular components.

flowchart TD
    A["🚀 Start Pipeline"] --> B["0️⃣ Template Init<br/>src/template/"]
    B --> C["1️⃣ Setup & Dependencies<br/>src/setup/"]
    C --> D["2️⃣ Tests<br/>src/tests/"]
    D --> E["3️⃣ GNN Discovery & Parsing<br/>src/gnn/"]
    E --> F["4️⃣ Model Registry<br/>src/model_registry/"]
    F --> G["5️⃣ Type Checking<br/>src/type_checker/"]
    G --> H["6️⃣ Validation<br/>src/validation/"]
    H --> I["7️⃣ Export<br/>src/export/"]
    I --> J["8️⃣ Visualization<br/>src/visualization/"]
    J --> K["9️⃣ Advanced Viz<br/>src/advanced_visualization/"]
    K --> L["1️⃣0️⃣ Ontology<br/>src/ontology/"]
    L --> M["1️⃣1️⃣ Rendering<br/>src/render/"]
    M --> N["1️⃣2️⃣ Execution<br/>src/execute/"]
    N --> O["1️⃣3️⃣ LLM Analysis<br/>src/llm/"]
    O --> P["1️⃣4️⃣ ML Integration<br/>src/ml_integration/"]
    P --> Q["1️⃣5️⃣ Audio<br/>src/audio/"]
    Q --> R["1️⃣6️⃣ Analysis<br/>src/analysis/"]
    R --> S["1️⃣7️⃣ Integration<br/>src/integration/"]
    S --> T["1️⃣8️⃣ Security<br/>src/security/"]
    T --> U["1️⃣9️⃣ Research<br/>src/research/"]
    U --> V["2️⃣0️⃣ Website<br/>src/website/"]
    V --> W["2️⃣1️⃣ Report<br/>src/report/"]
    W --> X["2️⃣2️⃣ MCP<br/>src/mcp/"]
    X --> Y["✅ Complete"]

    style A fill:#e1f5fe
    style G fill:#fff3e0,stroke:#ff9800,stroke-width:2px
    style Y fill:#e8f5e8,stroke:#4caf50

🎯 GNN Processing Workflow

flowchart TD
    subgraph "🧠 GNN Processing Workflow"
        A["📄 GNN File Input<br/>.md format"] --> B["🔍 Discovery & Parsing<br/>Extract sections"]
        B --> C["✅ Type Checking<br/>Validate syntax & structure"]
        C --> D["📊 Multi-Format Export<br/>JSON, XML, GraphML, Pickle"]
        
        C --> E["🎨 Visualization<br/>Generate graphs & matrices"]
        C --> F["🔄 Code Generation<br/>PyMDP, RxInfer & ActiveInference.jl templates"]
        
        F --> G["▶️ Simulation Execution<br/>Run generated code"]
        E --> H["🧠 LLM Analysis<br/>AI-powered insights"]
        
        D --> I["📚 Ontology Mapping<br/>Active Inference terms"]
        G --> J["📈 Results Analysis<br/>Performance metrics"]
        H --> K["🎵 Audio Generation<br/>SAPF, Pedalboard backends"]
        
        I --> L["🌐 Site Generation<br/>Static HTML reports"]
        J --> L
        K --> M["📊 Report Generation<br/>Comprehensive analysis"]
        
        L --> M
        M --> N["✨ Complete Analysis<br/>Multi-modal outputs"]
    end
    
    style A fill:#e1f5fe,stroke:#0277bd
    style C fill:#fff3e0,stroke:#f57c00
    style L fill:#e8f5e8,stroke:#388e3c

🏗️ Pipeline Architecture: Three-Tier Pattern

The pipeline follows a three-tier architectural pattern for maintainability and modularity:

main.py → Numbered Scripts (Thin Orchestrators) → Modular Scripts in Folders

🏛️ Architectural Components

Main Pipeline Orchestrator (src/main.py): Central coordinator that executes numbered scripts in sequence
Thin Orchestrators (src/0_template.py, src/1_setup.py, etc.): Minimal scripts that delegate to modules
Modular Scripts (src/template/, src/setup/, etc.): Core functionality implementation

📋 Current Status

✅ Compliant Scripts (7/23):

src/0_template.py → src/template/
src/1_setup.py → src/setup/
src/6_validation.py → src/validation/
src/7_export.py → src/export/
src/8_visualization.py → src/visualization/
src/10_ontology.py → src/ontology/
src/11_render.py → src/render/

🔄 Pending Refactoring (16/23):

src/2_tests.py → src/tests/ (8 functions)
src/3_gnn.py → src/gnn/ (1 function)
src/4_model_registry.py → src/model_registry/ (3 functions)
src/5_type_checker.py → src/type_checker/ (4 functions)
src/9_advanced_viz.py → src/advanced_visualization/ (10 functions)
src/12_execute.py → src/execute/ (14 functions)
And 10 more scripts...

📁 Example Structure

src/
├── main.py                          # Main pipeline orchestrator
├── 0_template.py                    # Thin orchestrator - imports from template/
├── template/                        # Modular template implementation
│   ├── __init__.py                 # Module exports and initialization
│   ├── processor.py                # Core template processing logic
│   └── mcp.py                      # Model Context Protocol integration
└── tests/
    └── test_template_integration.py # Tests for template module

📚 Documentation

For comprehensive architectural documentation, see:

src/template/README.md: Reference implementation and pattern documentation
ARCHITECTURE.md: Complete architectural guide

🚀 Running the Pipeline

Navigate to the project's root directory and execute:

python src/main.py [options]

🛠️ Key Pipeline Options

Option	Description	Default
`--target-dir DIR`	Target directory for GNN files	`src/gnn/examples`
`--output-dir DIR`	Directory to save outputs	`output/`
`--recursive`	Recursively process directories	`True`
`--skip-steps LIST`	Skip specific steps (e.g., "1,7")	None
`--only-steps LIST`	Run only specific steps	None
`--verbose`	Enable detailed logging	`False`
`--strict`	Enable strict type checking	`False`
`--estimate-resources`	Estimate computational resources	`True`

📋 View All Pipeline Options

python src/main.py --help

Additional specialized options:

--ontology-terms-file FILE: Path to ontology terms file
--llm-tasks LIST: Comma-separated LLM tasks
--llm-timeout: LLM processing timeout
--pipeline-summary-file FILE: Pipeline summary report path
--site-html-filename NAME: Generated HTML site filename
--duration: Audio duration for audio generation (default: 30.0)
--audio-backend: Audio backend to use (auto, sapf, pedalboard, default: auto)
--recreate-uv-env: Recreate UV environment
--dev: Install development dependencies

🛠️ Tools and Utilities

The GNN ecosystem includes several sophisticated tools to aid in model development, validation, and understanding. These tools are primarily invoked through the src/main.py pipeline script.

✅ Type Checker and Resource Estimator

The GNN Type Checker (pipeline step 4) helps validate GNN files and estimates computational resources.

🎯 Quick Usage

# Run only type checker
python src/main.py --only-steps 4 --target-dir path/to/gnn_files

# Include resource estimation
python src/main.py --only-steps 4 --estimate-resources --target-dir path/to/gnn_files

# Run full pipeline
python src/main.py --target-dir path/to/gnn_files

📊 Features

✅ Validation of required sections and structure
🔍 Type checking of variables and dimensions
🔗 Verification of connections and references
📋 Detailed error reports with suggestions for fixes
💾 Resource usage estimation and optimization recommendations

📁 Output Structure

When executed, the type checker writes to output/5_type_checker_output/:

output/5_type_checker_output/
├── type_check_results.json
├── type_check_summary.json
└── global_type_analysis.json

🎨 Visualization

GNN files can be visualized to create comprehensive graphical representations of models (pipeline step 6).

🎯 Usage

# Generate visualizations
python src/main.py --only-steps 6 --target-dir path/to/gnn_file.md

🖼️ Visualization Types

📊 Network Graphs: Model structure and dependencies
🎨 Matrix Heatmaps: A, B, C, D matrix visualizations
🧠 Ontology Diagrams: Active Inference relationship maps
📈 Performance Plots: Resource usage and timing analysis

🚀 Quick Start

New to GNN? Choose your learning path:

🎯 Choose Your Journey

⚡ Quick Demo (5 min): See GNN in action → 5-Minute Demo
🔬 I'm a Researcher: Theory-first approach → Research Path
💻 I'm a Developer: Code-first approach → Developer Path
🎓 I'm Learning: Structured curriculum → Academic Path

📚 Need guidance choosing? → Complete Learning Paths Guide

🛠️ Direct Installation (if you know what you want)

1️⃣ Prerequisites

Ensure you have Python 3.8+ installed:

python --version  # Should show 3.8 or higher

2️⃣ Clone Repository

git clone https://github.com/ActiveInferenceInstitute/GeneralizedNotationNotation.git
cd GeneralizedNotationNotation

3️⃣ Setup Environment

Run the setup pipeline step to configure dependencies:

python src/main.py --only-steps 1 --dev

This will:

✅ Create and configure virtual environment
📦 Install all required dependencies
🧪 Install development dependencies (with --dev)
✅ Validate system requirements

4️⃣ Run Your First Pipeline

Process the example GNN files:

python src/main.py --target-dir src/gnn/examples --verbose

5️⃣ Explore Results

Check the generated outputs in the output/ directory. The static site is under the numbered website folder:

ls -la output/
open output/20_website_output/website/index.html  # macOS
# or
xdg-open output/20_website_output/website/index.html  # Linux

🆘 Need Help?

🔍 Common Issues & Solutions

🐍 Python Version Issues

# Check Python version
python --version
# If < 3.8, install Python 3.8+ from python.org

📦 Dependency Issues

# Force reinstall dependencies
uv run python src/main.py --only-steps 2 --recreate-uv-env --dev

🔧 Pipeline Failures

# Run with verbose logging
python src/main.py --verbose
# Check specific step
python src/main.py --only-steps 4 --verbose

💾 Disk Space Issues

# Check available space
df -h
# Clean output directory
rm -rf output/*

🔗 Get Support:

📖 Documentation: See Documentation section below
🐛 Known Issues: Check troubleshooting guide
💬 Community: Open an issue on GitHub
🚀 Quick Fix: Try python src/main.py --only-steps 2 --dev first

📖 Documentation

Comprehensive documentation is organized in the doc/ directory. Here are the key resources:

📚 Core Documentation

Document	Description
GNN Overview	High-level introduction to GNN
Syntax Guide	Detailed GNN syntax specification
File Structure	GNN file organization guide
Quick Start Tutorial	Step-by-step beginner guide
Comprehensive Project Docs	Full architecture, pipeline, and dataflow diagrams
Architecture Guide	Implementation-oriented architecture and extension patterns
Machine-Readable Indices	API index and generator

🎯 Specialized Guides

Topic	Documentation
🧠 Active Inference	About GNN
🤖 LLM Integration	LLM & Neurosymbolic AI
📊 Implementation	Implementation Guide
🛠️ Tools	Tools & Resources
📄 Research Paper	Academic Paper Details

🎯 Integration Guides

Platform	Documentation
🐍 PyMDP	PyMDP Integration
🔬 RxInfer.jl	RxInfer Integration
🧠 ActiveInference.jl	ActiveInference.jl Integration
📡 MCP	Model Context Protocol
🧮 SymPy	Mathematical Processing
🔄 DisCoPy	Categorical Diagrams

🧩 Application Examples

Domain	Examples
🧠 Cognitive Phenomena	Cognitive Models
🎯 Templates	Model Templates
📋 Configuration	Configuration Examples

🎯 Examples

Explore practical GNN implementations and use cases:

📂 Example Files Location

📁 Primary Examples: src/gnn/examples/
📁 Cognitive Models: doc/cognitive_phenomena/
📁 Templates: doc/templates/

🔥 Featured Examples

Example	Description	Location
🎯 PyMDP POMDP Agent	Complete POMDP implementation	`src/gnn/examples/actinf_pomdp_agent.md`
🔬 RxInfer Hidden Markov Model	Probabilistic sequence modeling	`doc/archive/rxinfer_hidden_markov_model.md`
🧠 ActiveInference.jl Examples	Julia-based Active Inference models	`doc/activeinference_jl/actinf_jl_src/`
🤝 Multi-Agent System	Collaborative agent modeling	`doc/archive/rxinfer_multiagent_gnn.md`

🧠 Cognitive Phenomena Examples

Phenomenon	Model	Documentation
🎯 Attention	Attention mechanisms	`doc/cognitive_phenomena/attention/`
🧠 Consciousness	Global workspace theory	`doc/cognitive_phenomena/consciousness/`
💪 Cognitive Effort	Effort and control	`doc/cognitive_phenomena/effort/`
❤️ Emotion & Affect	Interoceptive emotion	`doc/cognitive_phenomena/emotion_affect/`
🎮 Executive Control	Task switching	`doc/cognitive_phenomena/executive_control/`

🏃‍♂️ Running Examples

# Process all examples
python src/main.py --target-dir src/gnn/examples

# Process specific example
python src/main.py --target-dir src/gnn/examples/actinf_pomdp_agent.md

# Process with full analysis
python src/main.py --target-dir src/gnn/examples --estimate-resources --verbose

⚡ Power User Tips

🚀 Advanced Usage Patterns

🔥 Quick Commands

# Full pipeline with all features
python src/main.py --verbose --estimate-resources --dev

# Type check only (fastest validation)  
python src/main.py --only-steps 4 --strict

# Visualization only (quick preview)
python src/main.py --only-steps 6

# Complete analysis for single file
python src/main.py --target-dir path/to/file.md --verbose

🎯 Pipeline Optimization

# Skip time-consuming steps for quick iteration
python src/main.py --skip-steps "11,12,13"

# Focus on core processing
python src/main.py --only-steps "1,4,5,6"

# Development workflow
python src/main.py --only-steps "2,3" --dev

📊 Output Management

# Custom output directory
python src/main.py -o /path/to/custom/output

# Timestamped outputs
python src/main.py -o "output/run_$(date +%Y%m%d_%H%M%S)"

🤝 Contributing

GNN is an evolving standard, and contributions are welcome! Here's how you can get involved:

🎯 Ways to Contribute

🐛 Report Issues: Found a bug? Open an issue
💡 Suggest Features: Have ideas? Start a discussion
📝 Improve Documentation: Help make our docs better
🧪 Add Examples: Share your GNN models
🔧 Code Contributions: Submit pull requests

📋 Contribution Guidelines

🍴 Fork the repository
🌿 Create a feature branch (git checkout -b feature/amazing-feature)
✅ Test your changes thoroughly
📝 Document your changes
💾 Commit with clear messages (git commit -m 'Add amazing feature')
📤 Push to your branch (git push origin feature/amazing-feature)
🔄 Submit a Pull Request

🛡️ Code of Conduct

Please read our Code of Conduct to understand the standards we maintain for our community.

📞 Getting Help

📖 Documentation: Check the docs first
💬 Discussions: Use GitHub Discussions
🐛 Issues: For bugs, use GitHub Issues
📧 Contact: Reach out to the maintainers

🙌 Recognition

All contributors will be recognized in our contributors list and release notes.

📄 License

This project is licensed under the MIT License. See the LICENSE.md file for full details.

📋 License Summary

✅ Commercial use permitted
✅ Modification permitted
✅ Distribution permitted
✅ Private use permitted
❗ License and copyright notice required

🔗 External Resources & Links

🌐 Active Inference Community

🏠 Active Inference Institute: activeinference.institute
💬 Community Discussions: Active Inference Discord

🛠️ Technical Resources

🐍 PyMDP Framework: github.com/infer-actively/pymdp
🔬 RxInfer.jl: github.com/biaslab/RxInfer.jl
🧠 ActiveInference.jl: github.com/ComputationalPsychiatry/ActiveInference.jl
📊 DisCoPy: github.com/oxford-quantum-group/discopy
📚 Research Papers: Active Inference on arXiv

Built with ❤️ by the Active Inference community

⬆️ Back to top

Name		Name	Last commit message	Last commit date
Latest commit History 227 Commits
.cursor_rules		.cursor_rules
doc		doc
input		input
output		output
scripts		scripts
src		src
.cursorrules		.cursorrules
.gitignore		.gitignore
ARCHITECTURE.md		ARCHITECTURE.md
CITATION.cff		CITATION.cff
CODE_OF_CONDUCT.md		CODE_OF_CONDUCT.md
COMPREHENSIVE_PIPELINE_ASSESSMENT.md		COMPREHENSIVE_PIPELINE_ASSESSMENT.md
CONTRIBUTING.md		CONTRIBUTING.md
DOCS.md		DOCS.md
DOCUMENTATION_ASSESSMENT.md		DOCUMENTATION_ASSESSMENT.md
LICENSE.md		LICENSE.md
README.md		README.md
SECURITY.md		SECURITY.md
SUPPORT.md		SUPPORT.md
matrix_analysis.png		matrix_analysis.png
pyproject.toml		pyproject.toml
pytest.ini		pytest.ini
requirements.txt		requirements.txt
uv.lock		uv.lock

License

ActiveInferenceInstitute/GeneralizedNotationNotation

Folders and files

Latest commit

History

Repository files navigation

GeneralizedNotationNotation (GNN)

📋 Table of Contents

🌟 Overview

📚 Initial Publication

🎯 Core Design Principles

🎯 Motivation and Goals

🚫 Current Challenges

🎯 Our Goals

✨ Key Features

🎭 The Triple Play Approach

📋 Structured File Format

🏗️ Project Architecture

📁 Directory Structure

⚙️ Processing Pipeline

🎯 GNN Processing Workflow

🏗️ Pipeline Architecture: Three-Tier Pattern

🏛️ Architectural Components

📋 Current Status

📁 Example Structure

📚 Documentation

🚀 Running the Pipeline

🛠️ Key Pipeline Options

🛠️ Tools and Utilities

✅ Type Checker and Resource Estimator

🎯 Quick Usage

📊 Features

📁 Output Structure

🎨 Visualization

🎯 Usage

🖼️ Visualization Types

🚀 Quick Start

🎯 Choose Your Journey

🛠️ Direct Installation (if you know what you want)

🆘 Need Help?

📖 Documentation

📚 Core Documentation

🎯 Specialized Guides

🎯 Integration Guides

🧩 Application Examples

🎯 Examples

📂 Example Files Location

🔥 Featured Examples

🧠 Cognitive Phenomena Examples

🏃‍♂️ Running Examples

⚡ Power User Tips

🤝 Contributing

🎯 Ways to Contribute

📋 Contribution Guidelines

🛡️ Code of Conduct

📞 Getting Help

🙌 Recognition

📄 License

📋 License Summary

🔗 External Resources & Links

🌐 Active Inference Community

🛠️ Technical Resources

About

Topics

Resources

License

Code of conduct

Contributing

Security policy

Uh oh!

Stars

Watchers

Forks

Releases 1

Packages 0

Uh oh!

Contributors 2

Uh oh!

Languages

Packages