Coding & Development

DIG (Dive into Graphs) is a comprehensive open-source library designed for graph deep learning research. Unlike basic graph deep learning libraries, DIG offers a unified testbed for advanced, research-oriented tasks such as graph generation, self-supervised learning on graphs, explainability of Graph Neural Networks, deep learning on 3D graphs, and graph out-of-distribution. It provides unified implementations of data interfaces, common algorithms, and evaluation metrics, allowing researchers to easily implement their own methods and compare them against baseline methods using common datasets and metrics without extensive effort. The library supports various research directions including Graph Augmentation and Fair Graph Learning, and is built on PyTorch Geometric (PyG).

DeepLearningFlappyBird is an open-source project that showcases the application of Deep Reinforcement Learning, specifically Deep Q-learning, to train an AI agent to play the game Flappy Bird. This project is based on the Deep Q Learning algorithm described in "Playing Atari with Deep Reinforcement Learning" and generalizes it to the Flappy Bird environment. It provides a practical, hands-on example for individuals interested in understanding and implementing deep learning algorithms within game environments. The project details the installation process, the architecture of the convolutional neural network used, and the training methodology, including preprocessing steps and hyperparameter annealing. It is an excellent resource for educational purposes and experimentation with AI.

DeepLearningPython is a GitHub repository that offers updated scripts from neuralnetworksanddeeplearning.com, specifically tailored for Python 3.5.2 and integrated with the Theano deep learning library, including CUDA support. This resource provides a practical foundation for individuals looking to learn and implement neural networks. The repository includes three distinct network implementations (network.py, network2.py, network3.py) from the original book, all runnable via a single testing file, test.py. This setup allows users to easily train and evaluate different network configurations, with examples and comments linking back to specific chapters of the book. It's an excellent tool for hands-on learning and experimentation in deep learning.

deep-learning-v2-pytorch is a comprehensive repository offering projects and exercises for Udacity's Deep Learning Nanodegree program. It features a collection of tutorial notebooks covering diverse deep learning topics, guiding users through the implementation of models such as convolutional networks, recurrent networks, and Generative Adversarial Networks (GANs). The resource also delves into other essential concepts like weight initialization and batch normalization. Beyond tutorials, it provides starting code for Nanodegree projects, which are typically reviewed by Udacity reviewers. This repository is ideal for students and learners looking to gain practical experience and deepen their understanding of deep learning with PyTorch.

deep_learning_curriculum offers an advanced, open-source curriculum designed for individuals seeking to understand the latest developments in deep learning, with a particular emphasis on large language model alignment. It is hosted on GitHub and is intended for those with a strong quantitative background who are already familiar with the fundamentals of deep learning. The curriculum is structured into nine chapters covering topics like Transformers, Scaling Laws, Optimization, Reinforcement Learning, and Alignment. Each chapter includes recommended reading, optional reading, and suggested exercises to facilitate hands-on learning. While challenging, it provides a comprehensive pathway for self-study or mentored learning in this rapidly evolving field.

deep-learning-bitcoin is an open-source project designed to analyze and predict Bitcoin price patterns through deep learning. It employs a unique approach by training models on raw pixel data, mimicking how experienced human traders interpret price charts. The project provides functionalities to download Bitcoin tick data, convert it into 5-minute Open High Low Close (OHLC) representations, and train AI models like AlexNet. Initial results show a 70% accuracy in predicting upward or downward price movements on a dataset of 20,000 samples. The project aims to scale to larger datasets, integrate more complex convolutional neural networks like Google LeNet, and incorporate bar volumes for enhanced prediction accuracy.

everything-claude-code is a comprehensive performance optimization system designed for AI agent harnesses, originating from an Anthropic hackathon winner. It goes beyond simple configurations, offering a complete system that includes skills, instincts, memory optimization, continuous learning, security scanning, and research-first development. The tool provides production-ready agents, skills, hooks, rules, and configurations that have evolved over months of intensive daily use in building real products. It supports a wide range of AI agent harnesses, including Claude Code, Codex, Cursor, OpenCode, and Gemini, making it a versatile solution for developers working with different platforms. The system emphasizes token optimization, memory persistence, and verification loops to ensure efficient and reliable AI agent operation.

Efficient-3DCNNs offers a PyTorch implementation of "Resource Efficient 3D Convolutional Neural Networks," complete with source code and pretrained models. This repository is designed for developers and researchers working with 3D CNNs, providing a foundation for building and evaluating efficient models. It includes implementations of popular architectures like 3D SqueezeNet, MobileNet, ShuffleNet, and their v2 versions, as well as 3D ResNet and ResNeXt models. The tool supports various datasets such as Kinetics, Jester, and UCF-101, and offers functionalities for training from scratch, resuming training, and fine-tuning with pretrained models. It also includes utilities for data preparation, augmentation, FLOPs calculation, and video accuracy assessment.

Efficient-Deep-Learning is a comprehensive GitHub repository dedicated to collecting recent methods for deep neural network compression and acceleration. It categorizes techniques into neural architecture re-design or search (NAS), pruning (including structured and unstructured), quantization, matrix/low-rank decomposition, and knowledge distillation (KD). The repository particularly focuses on pruning (with lottery ticket hypothesis or LTH as a sub-topic), KD, and quantization, offering a curated list of relevant papers and surveys dating back to the 1980s. It serves as a valuable resource for researchers and practitioners aiming to improve the efficiency, speed, and compactness of deep learning models.

GPT-NeoX is EleutherAI's open-source library designed for training large-scale language models on GPUs. It builds upon NVIDIA's Megatron Language Model and integrates techniques from DeepSpeed, along with novel optimizations. The project aims to be a centralized and accessible resource for advanced techniques in large-scale autoregressive language model training, accelerating research in the field. It uniquely supports a wide array of systems and hardware, including Slurm, MPI, and IBM Job Step Manager, and has been successfully deployed on platforms like AWS, CoreWeave, and ORNL Summit. Key features include distributed training with ZeRO and 3D parallelism, cutting-edge architectural innovations like rotary and alibi positional embeddings, and easy integration with the open-source ecosystem, such as Hugging Face libraries and experiment monitoring tools like WandB and TensorBoard. It is primarily intended for users looking to train models with billions of parameters from scratch.

GPT2 is an open-source implementation for training and using GPT-2 models, designed to support both GPUs and TPUs. While not the official OpenAI implementation, it aims to closely follow the original GPT-2 specifications. Users can download pretrained models like "117M", "PrettyBig", and "1.5B", or train their own models using custom datasets. The tool provides functionality for generating text from prompts, either directly via command line or from a file. It also includes scripts for generating datasets from sources like openwebtext or user-provided text files, with detailed instructions for configuration and input function creation. The implementation is highly configurable via JSON files, allowing users to define model parameters, training settings, and data paths.

Neuronika is a machine learning framework built entirely in Rust, emphasizing ease of use, rapid prototyping, and performance. At its core, Neuronika utilizes reverse-mode automatic differentiation, enabling the creation of dynamically changing neural networks with minimal effort and overhead through a lean, imperative, and define-by-run API. The framework leverages the power of the Rust language to offer an intuitive and efficient interface without the need for Foreign Function Interfaces (FFI). It supports GPU-accelerated primitives via CUDA, serialization with Serde, and transparent BLAS support for optimized matrix multiplication. Neuronika is currently in active development, with breaking changes expected as it evolves.

OperForce AI is an enterprise-grade AI platform designed to streamline the development and deployment of artificial intelligence applications. The platform specializes in providing solutions for agentic AI, enabling the creation of autonomous AI systems that can perform complex tasks. It also focuses on AI-driven business process automation, helping organizations optimize their workflows and operational efficiency through intelligent automation. OperForce AI aims to serve enterprise clients by offering robust tools and infrastructure for building and scaling AI initiatives, positioning itself as a key player in the agentic AI startup landscape.

Local-LLM-Comparison-Colab-UI is a resource for evaluating and comparing different Large Language Models (LLMs) that can be run on standard consumer hardware. The project provides a collection of Colab WebUI links, allowing users to easily access and experiment with various LLMs. Initially, the repository aimed to score models against GPT-4, but it has evolved to focus on providing direct access to newer, capable models for users to test firsthand. This approach acknowledges that the best model often depends on specific use cases, encouraging users to assess performance directly. The repository includes a wide range of models, from 7B to 34B parameters, with notes on their characteristics like censorship, roleplay capabilities, and specific functionalities such as coding or NSFW content.

NATSpeech is a comprehensive open-source framework for Non-Autoregressive Text-to-Speech (NAR-TTS) research and development. It offers official PyTorch implementations of advanced models like PortaSpeech (NeurIPS 2021) and DiffSpeech (AAAI 2022), facilitating high-quality and portable speech generation. The framework includes robust features such as data processing for NAR-TTS using Montreal Forced Aligner, a scalable training and inference system, and an efficient random-access dataset implementation. It's designed for technical users who want to explore and build upon state-of-the-art speech synthesis technologies, providing the necessary tools and code for experimentation and deployment.

Proximus Ada is the first Belgian center of excellence dedicated to Artificial Intelligence and Cybersecurity. It aims to provide tailor-made solutions to meet the evolving needs of society by leveraging the power of AI and cybersecurity. The center offers services in advanced artificial intelligence technologies to provide decisive knowledge for businesses and protects organizations from cyber threats with a team of cybersecurity experts. Additionally, Proximus Ada engages in social initiatives, collaborating with research centers and universities to enhance digital technology adoption in society. Its mission is to drive transformation through AI and Cybersecurity solutions for the Proximus Group and its partners.

LLM-Shearing is an open-source tool developed by Princeton NLP for accelerating language model pre-training through structured pruning. It offers base models like Sheared-LLaMA-1.3B and Sheared-Pythia-160m, as well as pruned models without continued pre-training and instruction-tuned models such as Sheared-LLaMA-1.3B-ShareGPT. The tool provides a codebase for pruning and continued pre-training algorithms, demonstrating that pruning strong base models is a cost-effective way to achieve powerful small-scale language models. It includes detailed instructions for installation, data preparation, model conversion, and sample scripts for pruning and continued pre-training, built upon MosaicML's Composer package.

Machine-Learning-with-Python is an open-source repository offering a comprehensive collection of practice and tutorial-style Jupyter notebooks. It covers a wide array of machine learning techniques, including regression, classification, clustering, and dimensionality reduction. The repository also delves into deep learning concepts, object-oriented programming for ML, and practical deployment examples. It's designed to help users understand and implement various ML algorithms using Python, with detailed examples and explanations. The notebooks utilize popular libraries such as NumPy, Pandas, Scikit-learn, TensorFlow, and Keras, making it an excellent resource for both learning and practical application in data science.

neurojs is an open-source JavaScript framework designed for deep learning and reinforcement learning applications within the browser environment. While it mainly focuses on reinforcement learning, it is versatile enough for various neural network-based tasks. The library includes practical examples and demos, such as a 2D self-driving car visualization, to showcase its capabilities. It supports advanced features like uniform and prioritized replay buffers, advantage-learning, and models such as deep-q-networks and actor-critic (via deep-deterministic-policy-gradients). neurojs also allows for binary import and export of network configurations, including weights, and is built for high performance. However, development on neurojs is no longer actively maintained, with the recommendation to use more general frameworks like TensorFlow-JS.

Norse is an open-source deep learning library designed for spiking neural networks (SNNs) within the PyTorch framework. It aims to leverage the advantages of bio-inspired neural components, which are inherently sparse and event-driven, offering a fundamental difference from traditional artificial neural networks. By expanding PyTorch with these specialized primitives, Norse provides a modern and robust infrastructure for researchers and developers. The library includes various neuron models, synapse dynamics, encoding and decoding algorithms, and dataset integrations, making it a comprehensive tool for modeling scalable experiments. Norse is actively used in research and is optimized for performance, scaling efficiently from single laptops to HPC clusters.

Siemba offers a full-stack Continuous Threat Exposure Management (CTEM) platform designed to continuously identify, prioritize, and remediate threats across an organization's attack surface. It leverages an AI-powered assistant to facilitate faster decision-making, smarter prioritization of risks, and automated threat response. Key features include External Attack Surface Management (EASM), AI-driven Vulnerability Assessments (GenVA), and AI-driven Dynamic Application Security Testing (GenPT) with PenTest as a Service (PTaaS). The platform aims to enhance developer efficiency, align security teams, reduce external attack surfaces, and simplify compliance with one-click reporting. Siemba is recognized by Gartner for its CTEM capabilities.

Truelink.ai is an AI and VPN application designed to offer comprehensive real-time protection against various online threats. It integrates advanced artificial intelligence capabilities with robust VPN functionality, creating a secure browsing environment for its users. The tool aims to safeguard individuals and businesses by providing a dual-layered defense system. While specific features beyond AI and VPN integration are not detailed, the core offering focuses on enhancing online security through intelligent threat detection and encrypted internet access. This combination positions Truelink.ai as a solution for those seeking a proactive approach to digital safety.

onnx-go offers Go developers the capability to integrate pre-trained neural networks into their applications. It acts as an interface to the Open Neural Network Exchange (ONNX) format, enabling the decoding of ONNX binary models into a computation backend. This tool is particularly useful for adding machine learning capabilities to Go code without requiring specialized data science skills or being tied to a specific framework. While the implementation of the ONNX spec is partial for import and non-existent for export, it supports various backends like Gorgonia. The project is actively maintained by Orama and provides utilities to run models from the ONNX model zoo, making it a valuable resource for Go-based AI development.

PromptHub is an open-source, local-first AI prompt and skill management tool designed to streamline prompt engineering workflows. It offers robust features for creating, editing, and organizing prompts with support for folders, tags, and automatic version control, including the ability to view, compare, and roll back changes. Users can leverage template variables for dynamic prompt generation and quickly access frequently used prompts. A standout feature is its skill management system, which includes a skill store with pre-built skills and one-click installation to over 15 mainstream AI coding tools like Claude Code and Cursor. PromptHub also provides AI testing capabilities for comparing various models and supports local data storage with WebDAV cloud synchronization, ensuring privacy and data security. It is available as a desktop application for macOS, Windows, and Linux, and also offers a self-hosted web version.