Open Source BSD Neural Network Libraries

Neuroph is lightweight Java Neural Network Framework which can be used to develop common neural network architectures. Small number of basic classes which correspond to basic NN concepts, and GUI editor makes it easy to learn and use.

Alpa is a system for training and serving large-scale neural networks. Scaling neural networks to hundreds of billions of parameters has enabled dramatic breakthroughs such as GPT-3, but training and serving these large-scale neural networks require complicated distributed system techniques. Alpa aims to automate large-scale distributed training and serving with just a few lines of code.

SVoice is a PyTorch-based implementation of Facebook Research’s study on speaker voice separation as described in the paper “Voice Separation with an Unknown Number of Multiple Speakers.” This project presents a deep learning framework capable of separating mixed audio sequences where several people speak simultaneously, without prior knowledge of how many speakers are present. The model employs gated neural networks with recurrent processing blocks that disentangle voices over multiple computational steps, while maintaining speaker consistency across output channels. Separate models are trained for different speaker counts, and the largest-capacity model dynamically determines the actual number of speakers in a mixture. The repository includes all necessary scripts for training, dataset preparation, distributed training, evaluation, and audio separation.

Fast Artificial Neural Network Library is a free open source neural network library, which implements multilayer artificial neural networks in C with support for both fully connected and sparsely connected networks. Cross-platform execution in both fixed and floating point are supported. It includes a framework for easy handling of training data sets. It is easy to use, versatile, well documented, and fast. Bindings to more than 15 programming languages are available. An easy to read introduction article and a reference manual accompanies the library with examples and recommendations on how to use the library. Several graphical user interfaces are also available for the library.

CRF-RNN is a deep neural architecture that integrates fully connected Conditional Random Fields (CRFs) with Convolutional Neural Networks (CNNs) by reformulating mean-field CRF inference as a Recurrent Neural Network. This fusion enables end-to-end training via backpropagation for semantic image segmentation tasks, eliminating the need for separate, offline post-processing steps. Our work allows computers to recognize objects in images, what is distinctive about our work is that we also recover the 2D outline of objects. Currently we have trained this model to recognize 20 classes. This software allows you to test our algorithm on your own images – have a try and see if you can fool it, if you get some good examples you can send them to us. CRF-RNN has been developed as a custom Caffe layer named MultiStageMeanfieldLayer. Usage of this layer in the model definition prototxt file looks the following. Check the matlab-scripts or the python-scripts folder for more detailed examples.

Neuroph OCR - Handwriting Recognition is developed to recognize hand written letter and characters. It's engine derived's from the Java Neural Network Framework - Neuroph and as such it can be used as a standalone project or a Neuroph plug in.

nn_vis is a minimalist visualization tool for neural networks written in Python using OpenGL and Pygame. It provides an interactive, graphical representation of how data flows through neural network layers, offering a unique educational experience for those new to deep learning or looking to explain it visually. By animating input, weights, activations, and outputs, the tool demystifies neural network operations and helps users intuitively grasp complex concepts. Its lightweight codebase is great for customization and teaching purposes.

Neural Processes (NPs) is a collection of interactive Jupyter/Colab notebook implementations developed by Google DeepMind, showcasing three foundational probabilistic machine learning models: Conditional Neural Processes (CNPs), Neural Processes (NPs), and Attentive Neural Processes (ANPs). These models combine the strengths of neural networks and stochastic processes, allowing for flexible function approximation with uncertainty estimation. They can learn distributions over functions from data and efficiently make predictions at new inputs with calibrated uncertainty — making them useful for few-shot learning, Bayesian regression, and meta-learning. Each notebook includes theoretical explanations, key building blocks, and executable code that runs directly in Google Colab, requiring no local setup. Implementations rely only on standard dependencies such as NumPy, TensorFlow, and Matplotlib, and provide visualizations of model performance.

Lightweight backpropagation neural network in C. Intended for programs that need a simple neural network and do not want needlessly complex neural network libraries. Includes example application that trains a network to recognize handwritten digits.

NeuronDotNet is a neural network engine written in C#. It provides an interface for advanced AI programmers to design various types of artificial neural networks and use them.

Feedforward Neural Network writen in C++ serial and parallelized in TBB library. Also using Autotune library for best parallel performance.

Amygdala is a C++ spiking neural network library. It includes several neuron models, SMP support and facilities for developing SNNs with genetic algorithms. Support for running Amygdala neural networks on workstation clusters and MPPs is also under way

Do you want a neural network OO class? With documentation? Do not go further... You found it! you have here a complete library of different neural network in an OO encapsulation. Starting from adaline, back propagation, Kohonen

ANNJ, Another Neural Network for Java is a neural network framework for the Java programming language. It is still in an early development stage, currently supporting only feed-forward type networks, but will soon be able to handle many other types.

A java based neural network framework. The Auratus network is built around an XML messaging system, allowing for a complete MVC design. Additionally, Auratus networks are constructed and at the node/edge level, allowing for advanced topologies.

The subject of this project is development of the Intrusion Detection System (IDS) with Artificial Neural Network (ANN). This model will allow the implementation of a system capable to analyze and to identify possible intrusions, based on the method of an

Cluster Networks are a new style of neural simulation / neural network modeling, that models networks of neural populations ("clusters") that transform and transmit information using precisely-timed, graded bursts ("pulses" or "volleys") of firing.

compare_gan is a research codebase that standardizes how Generative Adversarial Networks are trained and evaluated so results are comparable across papers and datasets. It offers reference implementations for popular GAN architectures and losses, plus a consistent training harness to remove confounding differences in optimization or preprocessing. The library’s evaluation suite includes widely used metrics and diagnostics that quantify sample quality, diversity, and mode coverage. With configuration-driven experiments, you can sweep hyperparameters, run ablations, and log results at scale. The goal is to turn GAN experimentation into a disciplined, repeatable process rather than a patchwork of scripts. It also provides baselines strong enough to serve as starting points for new ideas without re-implementing the world.

CoreNet is Apple’s internal deep learning framework for distributed neural network training, designed for high scalability, low-latency communication, and strong hardware efficiency. It focuses on enabling large-scale model training across clusters of GPUs and accelerators by optimizing data flow and parallelism strategies. CoreNet provides abstractions for data, tensor, and pipeline parallelism, allowing models to scale without code duplication or heavy manual configuration. Its distributed runtime manages synchronization, load balancing, and mixed-precision computation to maximize throughput while minimizing communication bottlenecks. CoreNet integrates tightly with Apple’s proprietary ML stack and hardware, serving as the foundation for research in computer vision, language models, and multimodal systems within Apple AI. The framework includes monitoring tools, fault tolerance mechanisms, and efficient checkpointing for massive training runs.

The DAMNED simulator is a Distributed And Multithreaded Neural Event-Driven simulation framework designed for Distributed Discrete Event Simulations, actually well suited for Large Scale Spiking Neural Network simulations.

DeepDream is a small, educational repository that accompanies Google’s original “Inceptionism” blog post by providing a runnable IPython/Jupyter notebook that demonstrates how to “dream” through a convolutional neural network. The notebook shows how to take a trained vision model and iteratively amplify patterns the network detects, producing the hallmark surreal, hallucinatory visuals. It walks through loading a pretrained network, selecting layers and channels to maximize, computing gradients with respect to the input image, and applying multi-scale “octave” processing to reveal fine and coarse patterns. The code is intentionally compact and exploratory, encouraging users to tweak layers, step sizes, and scales to influence the aesthetic. Although minimal, it illustrates important concepts like feature visualization, activation maximization, and the effect of different receptive fields on the final image.

FAIRChem is a unified library for machine learning in chemistry and materials, consolidating data, pretrained models, demos, and application code into a single, versioned toolkit. Version 2 modernizes the stack with a cleaner core package and breaking changes relative to V1, focusing on simpler installs and a stable API surface for production and research. The centerpiece models (e.g., UMA variants) plug directly into the ASE ecosystem via a FAIRChem calculator, so users can run relaxations, molecular dynamics, spin-state energetics, and surface catalysis workflows with the same pretrained network by switching a task flag. Tasks span heterogeneous domains—catalysis (OC20-style), inorganic materials (OMat), molecules (OMol), MOFs (ODAC), and molecular crystals (OMC)—allowing one model family to serve many simulations. The README provides quick paths for pulling models (e.g., via Hugging Face access), then running energy/force predictions on GPU or CPU.

ffnet is a fast and easy-to-use feed-forward neural network training solution for python. Many nice features are implemented: arbitrary network connectivity, automatic data normalization, very efficient training tools, network export to fortran code. Now ffnet has also a GUI called ffnetui.

Foad is an open source software which receive an EKG Signal from scanner, WFDB database or heart sensors. Finding patient disease started by taking Fourier transform (FFT) from input signal and extract a single cycle. Based on some heuristic algorithm the most important feature like P , Q , R , S , T captured and feed to trained neural network. and so the final decision made by CNN library. As mentioned before this software also capable do some image processing on scanned paper to lower the final costs.

The aim of GUINNEA (Graphical User Interfaced Neural Network Architecture) is to develop a comfortable and high-featured neural net simulator which is highly configurable and flexible. It will support many neural nets and visualization features for those