Directory
Published directory entries matching your search.
A course by Andrej Karpathy on building neural networks from scratch in code, starting with backpropagation basics and building up to modern deep neural networks like GPT. Prerequisites are solid Python programming and intro-level math such as derivatives and gaussians.
Brainstorm is a GitHub repository for a neural network library developed by IDSIA. It is designed to be fast, flexible, and fun for neural network development, with documentation and support resources available for users implementing the library.
NN Visual is an interactive web-based neural network visualizer. It lets users configure hidden layers, select datasets such as auto_mpg, iris, xor, or mnist, and train a network while observing forward and backpropagation step by step. The tool displays training loss, mean absolute error, weight distributions, and node activation brightness. Written explanations accompany the visualizations to teach concepts like gradient descent, weights, biases, and activation functions without requiring code.
Garage is a toolkit for developing and evaluating reinforcement learning algorithms, with an accompanying library of implementations built using it. The toolkit offers modular tools including composable neural network models, replay buffers, high-performance samplers, an experiment definition interface, reproducibility tools, logging including TensorBoard, checkpointing and resuming, and environment interfaces.
Animated AI is a project of animations and instructional videos about neural networks. Its topics cover convolution basics, padding, stride, grouped and depthwise-separable convolution, and pixel shuffle. It is licensed under the MIT License.
Antigravity Link is a browser-based mobile companion for Google's Antigravity IDE that mirrors active AI chat sessions on a phone over the local network. It also exposes a local HTTP API and an MCP server so agents and external tools can drive Antigravity sessions programmatically.
An AWS program offering students and educators no-cost access to cloud computing skills training developed by AWS experts, plus career pathways for students. It provides a library of free, self-paced online training and hands-on labs covering storage, compute, networking, databases and cloud operations. Learners as young as 13 can register with an email address, and learners 18 and older can access a job board.
Brain.js is a GitHub-hosted JavaScript library for building and running neural networks. It provides GPU-accelerated neural network functionality compatible with both web browsers and Node.js environments, allowing users to implement machine learning models using JavaScript across different runtime platforms.
Neural networks, regression and feature learning in Clojure.
FANN (Fast Artificial Neural Network) binding.
Course materials and notes for the Stanford class CS231n: Deep Learning for Computer Vision. The notes accompany the Stanford CS class and cover topics organized in modules. Spring 2026 assignments are listed.
deeplearn-rs is a Rust library for neural networks and deep learning, described by its author as mostly a proof of concept with a changing API. Implemented models include MNIST handwritten digit recognition and a char-rnn using LSTM, alongside layers such as fully connected, LSTM, softmax, and loss functions, with SGD and RMSProp optimizers. The repository was archived by its owner on Mar 8, 2024, and is read-only.
dl4clj is a Clojure port of deeplearning4j. It provides stable features with tests, a neural networks DSL, early stopping, training, transfer learning and evaluation. The project is not yet released to Clojars.
go-dnn is a GitHub repository that provides deep neural network functionality for the Golang programming language, powered by MXNet. The repository notes an updated version of the project is hosted at a separate GitHub location.
gobrain is an open-source GitHub repository maintained by the goml organization. It implements neural networks in the Go programming language. The repository includes documentation for getting started, recurrent neural network features, a changelog of updates, and project background information.
Greywall is a container-free, deny-by-default sandbox designed for AI coding agents. It provides kernel-enforced filesystem, network, and syscall isolation for Linux and macOS operating systems, and is hosted as a public repository on GitHub.
igraph is a collection of network analysis tools emphasizing efficiency, portability and ease of use. It is open source and free, and can be programmed in R, Python, Mathematica and C/C++. Implementations include the igraph R package, python-igraph, IGraph/M and the igraph C library.
A beginner-friendly guide on using Keras to implement a simple neural network in Python. It walks through MNIST handwritten digit classification, covering setup, data preparation, building the model, compiling, training, testing and using the model. It is intended for complete beginners to Keras who have a basic background knowledge of neural networks.
Simple API for Neural Network. Better for image processing with CPU/GPU + Transfer Learning.
LAION, the Large-scale Artificial Intelligence Open Network, is a non-profit organization that provides datasets, tools and models to support open machine learning research. Its offerings include the LAION-400M dataset of 400 million English image-text pairs, LAION-5B of 5.85 billion multilingual CLIP-filtered pairs, and the CLIP H/14 vision transformer model.
Collection of infrastructure and tools for research in neural network interpretability.
A repository of material for the exercise classes of a Big Data Analytics master course. Hands-on sessions cover Spark and its machine learning library, Map/Reduce implementations of linear regression and alternating least squares, streaming analytics with Recursive Least Squares, and image classification with convolutional neural networks. Material is provided as Jupyter notebooks.
Neural networks (boltzmann machines, feed-forward and recurrent nets), Gaussian Processes.
It implemented multi-perceptrons neural network (ニューラルネットワーク) based on Back Propagation Neural Networks (BPN) and designed unlimited-hidden-layers.
