self-organizing-maps

torchsom: The Reference PyTorch Library for Self-Organizing Maps (JMLR 2026). GPU-accelerated SOMs with a scikit-learn API, clustering and visualization.

Compendium for "A Scalable Clustering-Based Method for Vegetation Mapping in Large Areas Using Satellite Image Time Series" — SOM + HCA/DTW clustering pipeline for SITS, applied to secondary-vegetation classification in TerraClass Cerrado 2024.

FormicaX: Rust library with clustering algorithms like K-Means, DBSCAN, and GMM, FormicaX delivers efficient, adaptable insights for trading applications. Inspired by the collaborative and resilient nature of ants (Formica), it offers a modular, high-performance framework for developers and data scientists.

Topology-preserving manifold discovery for high-dimensional sparse signal spaces using Self-Organizing Maps (SOM). Focuses on non-convex quantization error minimization and structural visualization.

🔍 Advanced fraud detection system using Self-Organizing Maps for anomaly detection and Neural Networks for classification

Deep Learning codes

Unsupervised Clustering and Ensemble Decision Strategies in Cryptocurrency Trading: A SOM-Based Hybrid Model for Signal Generation

Cartes auto-organisatrices de Kohonen (SOM) en Python, étude d'hyperparamètres (eta, sigma, N) · Projet IA Polytech Lyon

A hybrid deep learning system combining Unsupervised SOM and Supervised Deterministic ANN for advanced financial fraud detection.

Texture Patch Clustering

Neural networks & deep learning (TC2035): perceptrons, MLPs, CNNs (LeNet-5), SOM/LVQ competitive learning and a Vision Transformer on Fashion-MNIST - with trained-model artifacts (87% test acc).