6 dépôts
Standardized rules for exchanging data via plain-text streams between processes.
Distinct from Communication Protocols and Standards: The candidates focus on low-level I/O streams or high-level network standards, not the specific text-protocol contract used by Git helpers.
Explore 6 awesome GitHub repositories matching networking & communication · Text-Based Communication Protocols. Refine with filters or upvote what's useful.
This project is a multi-language compiler collection and cross-platform toolchain used to translate source code from various programming languages into optimized machine code for different hardware architectures. It provides a suite of tools including an optimizing compiler backend, a machine code generator, and a comprehensive runtime library suite that implements necessary execution environments and support functions. The system utilizes a multi-pass compilation pipeline and pluggable language front-ends to process source code into intermediate representations. It distinguishes itself throu
Uses a text-encoded packet protocol for structured communication between the compiler and builder.
This project is a Git credential helper that automates the storage and retrieval of authentication secrets for remote repository operations. It functions as an OAuth token manager and an operating system vault storage interface to ensure authentication secrets are encrypted at rest. The tool acts as a cross-platform authentication broker, enabling the sharing of secure credentials between a host operating system and a Linux subsystem. It also serves as an enterprise proxy gateway, routing authentication traffic through corporate proxy servers to reach restricted repository endpoints. The sys
Communicates with Git clients using a standardized text-based input-output stream protocol.
Beanstalkd is a distributed work queue server designed for asynchronous job processing. It functions as a standalone server that distributes background tasks between producers and consumers to improve application responsiveness and throughput. The system organizes tasks using numeric priority levels to ensure critical work is processed first. It manages the job lifecycle through discrete states and uses a simple line-based text protocol over TCP for communication. To ensure reliability, the server persists job data to a sequential disk log, allowing the queue state to be recovered after a sy
Implements a simple line-based text communication format over TCP for low-overhead interactions.
Ce projet est un framework d'apprentissage par renforcement et un moteur d'IA de jeu conçu pour entraîner des agents antagonistes dans des jeux au tour par tour à deux joueurs. Il implémente une boucle d'entraînement qui utilise le self-play et la recherche arborescente de Monte Carlo pour produire des réseaux de neurones capables de prédire la force du plateau et les probabilités de coup. Le système découple le moteur d'apprentissage par renforcement des règles de jeu spécifiques via une interface de logique de jeu abstraite, permettant la définition de règles de jeu personnalisées, de conditions de victoire et de représentations de plateau. Il prend en charge l'intégration avec divers frameworks d'apprentissage profond pour servir de fonctions de politique et de valeur et inclut une architecture de réseau de neurones à double tête pour prédire à la fois les résultats attendus et les distributions de coups légaux. Le framework couvre un large éventail de capacités, notamment l'encodage de l'état du jeu, la gestion des points de contrôle des modèles et le suivi des métriques d'entraînement. Il fournit des outils pour évaluer la performance des agents via des simulations de tournois, des interfaces humain-agent et un protocole textuel pour se connecter à des moteurs de jeu externes. Les modèles entraînés peuvent être exportés vers un format compatible JavaScript pour une utilisation dans des environnements de déploiement basés sur le web.
Uses a standardized plain-text protocol to communicate moves and state between the agent and external engines.
Minigo is a TensorFlow-based reinforcement learning engine designed to master the game of Go. It functions as a comprehensive system for training neural networks to predict board policies and game outcomes, utilizing a model trainer to generate self-play data and optimize weights. The project is distinguished by its ability to perform large-scale game simulations using Kubernetes to distribute worker nodes across CPU, GPU, and TPU hardware. It employs a Monte Carlo Tree Search implementation to identify optimal moves and supports specialized hardware acceleration, including inference on Edge
Communicates engine state and move suggestions using a standardized text-based protocol.
MPD is a headless music server daemon that indexes audio libraries and streams music to local or remote outputs. It functions as a music library manager and network audio streamer, providing a remote audio control protocol that allows external clients to manage playback, playlists, and database queries. The system acts as a multiroom audio coordinator, synchronizing audio distribution across multiple networked clients and hardware devices. It supports a variety of remote management capabilities, including a dedicated control API and the ability to broadcast audio streams over network protocol
Provides a request-response text-based protocol for remote management of the music library and playback.