Autoencoder
Create a GPU-accelerated autoencoder network
Production-grade deep learning autoencoder for keystroke dynamics authentication
Library to build and run auto-encoder neural network powered by genetic algorithm or GPU.
model image client decoder
Node to Train AI Networks
model image client decoder
A simple Auto Encoder typescript library for experimentation and dimensionality reduction. Supports automatic scaling.
Run Keras models in the browser, with GPU support using WebGL
This small library has no dependencies and adds interfaces and classes that you can use to encode and decode your own classes to a JSON compatible structure. This enables you to build powerful API's and storage that can automigrate to newer versions on th
Dimensionality reduction methods
ConvNetJS is a Javascript implementation of Neural networks.
AI-powered workflow optimization using TensorFlow.js and YAWL patterns
ConvNetJs -- Version 2
TypeScript SDK for MachinaFi — Autonomous AI Agents on Base L2
A node.js library for locally synthesizing text to speech using VITS
Run Keras models in the browser, with GPU support using WebGL
Predictive systems with ML-based anomaly detection, maintenance, forecasting, and risk assessment
Simple to use neural network implementation in pure TypeScript with GPU support.
Extract features | Compress data | Deep network variants
Extract features | Compress data | Deep network variants
State-of-the-art clustering algorithms for Rust - surpassing scikit-learn, HDBSCAN, and RAPIDS cuML
Unsupervised learning and clustering algorithms for ToRSh, powered by SciRS2
Data transformation module for SciRS2 (scirs2-transform)
surge synthesizer -- wavetable oscillator
Pure Rust system call tracer with source-aware correlation for Rust binaries
Anomaly detection primitives for OxiCUDA — DeepSVDD, AE/VAE reconstruction, LOF, COPOD, isolation scoring, statistical methods, ensemble
Neural network building blocks module for SciRS2 (scirs2-neural) - Minimal Version
ST-EEGFormer EEG Foundation Model — inference in Rust with Burn ML
Time series analysis components for ToRSh - powered by SciRS2
Vision Transformer models and building blocks for Rust using tch.
Terminal based machine learning training monitor
HVAC diagnostic and fault-detection models built on the Axonml ML framework