This package provides an easy and modular way to build and train simple or complex neural networks using Torch:

• Modules are the bricks used to build neural networks. Each are themselves neural networks, but can be combined with other networks using containers to create complex neural networks:
  • Module: abstract class inherited by all modules;
  • Containers: composite and decorator classes like Sequential, Parallel, Concat and NaN;
  • Transfer functions: non-linear functions like Tanh and Sigmoid;
  • Simple layers: like Linear, Mean, Max and Reshape;
  • MLP: comprehensive Multi-Layer Perceptron module for building flexible feedforward neural networks;
  • Table layers: layers for manipulating tables like SplitTable, ConcatTable and JoinTable;
  • Convolution layers: Temporal, Spatial and Volumetric convolutions;
• Criterions compute a gradient according to a given loss function given an input and a target:
  • Criterions: a list of all criterions, including Criterion, the abstract class;
  • MSECriterion: the Mean Squared Error criterion used for regression;
  • ClassNLLCriterion: the Negative Log Likelihood criterion used for classification;
• Additional documentation:
  • Overview of the package essentials including modules, containers and training;
  • Training: how to train a neural network using StochasticGradient;
  • Testing: how to test your modules.
  • Experimental Modules: a package containing experimental modules and criteria.
• LLM and Learning Resources:
  • LLM Implementation: Neuro-symbolic LLM with dynamic learning for niche construction
  • Interesting Learnings: Deep dive into key insights and patterns from the codebase
  • Quick Reference: Concise guide to the most interesting learnings
  • Interactive Demo: Hands-on demonstration of key concepts
• Meta-Learning Resources:
  • 🔬 Structural Inversion: 🔬 META-COGNITIVE DISCOVERY - How code architecture and conceptual understanding exhibit inverse organizational principles
  • 🌟 Five Loops Complete Guide: 🌟 ULTIMATE INTEGRATION - All five learning loops unified from implementation to ontology
  • 🔮 Quintuple-Loop Ontology: 🔮 THE DEEPEST LEVEL - Exploring the ground of being and the nature of existence that enables learning
  • ⚡ Quintuple-Loop Quick Reference: ⚡ One-page guide to Loop 5 - when to use and key insights
  • 📖 Quintuple-Loop Summary: Executive summary of the Loop 5 discovery - being and knowing unified
  • 💫 Quintuple-Loop Example: Code demonstrating ontological awareness in practice
  • Four Loops Visual Guide: Complete visual guide to all four learning loops with practical examples
  • Four Loops Quick Reference: One-page reference card for rapid loop identification and decision-making
  • Learning Map: Visual guide to navigating all learning resources
  • Three Loops Quick Guide: Practical reference for choosing and using learning loops
  • Learning Evolution: Synthesis showing the journey through all three learning loops
  • Triple-Loop Learning: Examining the learning process itself - how we learn to learn
  • Quadruple-Loop Epistemology: Deep philosophical inquiry into the nature of understanding and knowledge itself
  • Quadruple-Loop Summary: Executive summary of the loop 4 discovery and its implications
  • Meta-Learning Reflection: Double-loop learning analysis questioning what "interesting" means
  • Double-Loop Insights: Actionable wisdom from questioning assumptions
  • Learning Loops Comparison: Visual guide comparing single-loop vs. double-loop learning
  • Double-Loop Field Guide: Practical application guide with templates and workflows
  • Double-Loop Summary: Executive summary of the double-loop learning journey