6 repos
Hardware Abstraction Layers — Artificial Intelligence & Machine Learning
We curate 6 GitHub repositories matching artificial intelligence & machine learning · Hardware Abstraction Layers. Refine with filters or upvote what's useful.
Hardware Abstraction Layers — Artificial Intelligence & Machine Learning
- torvalds/linux
torvalds/linux
217,986The Linux kernel is a monolithic operating system kernel that serves as the primary interface between computer hardware and software applications. It provides the foundational infrastructure for managing system resources, including memory allocation, process scheduling, and synchronization primitives. The project includes comprehensive support for diverse storage architectures through its filesystem suite and manages complex networking, virtualization, and power management subsystems. Beyond core system management, the kernel offers extensive frameworks for hardware interaction, covering input devices, audio, sensors, and various bus communication protocols. It incorporates diagnostic tools for system observability, security mechanisms for integrity protection, and a kernel-level virtual machine for sandboxed execution. The project maintains stability through defined interface guarantees and supports modular development, including integrated support for memory-safe programming.
C - tensorflow/tensorflow
tensorflow/tensorflow
193,864TensorFlow is a comprehensive machine learning framework designed for the construction, training, and deployment of complex mathematical models. It utilizes a graph-based execution model that represents operations as directed acyclic graphs, enabling automatic differentiation and efficient parallel processing. The system provides high-level interfaces for defining neural network architectures, alongside a robust engine for managing multidimensional array structures and tensor mathematics. The framework distinguishes itself through a scalable distributed runtime that orchestrates workloads across heterogeneous hardware accelerators and decentralized network nodes. It employs deferred-execution symbolic graphs to perform graph-level optimizations, fusion, and ahead-of-time kernel compilation for specific hardware architectures. To ensure consistent performance across production environments, it features a standardized serialization format for model graphs and specialized tools for model serving, quantization, and compression. Beyond core training capabilities, the platform includes a high-throughput data ingestion engine that supports asynchronous, multi-threaded pipelines to prevent bottlenecks. It also offers extensive support for hardware abstraction, allowing for pluggable device integration and containerized acceleration. The ecosystem is rounded out by utilities for data validation, federated learning, and specialized modeling tasks, providing a complete toolchain for moving models from research into high-availability production environments.
C++deep-learningdeep-neural-networksdistributed - godotengine/godot
godotengine/godot
106,855Godot is a comprehensive, node-based game engine designed for building interactive 2D and 3D applications. It provides an integrated development environment that utilizes a hierarchical scene system to organize objects, propagate spatial transformations, and manage lifecycle events. The engine functions as a cross-platform development suite, allowing developers to author, test, and export software to desktop, mobile, and web environments from a single, unified codebase. The engine distinguishes itself through a modular, component-based architecture that relies on signals-based decoupling for event-driven communication between objects. It features a server-side rendering architecture that separates high-level scene logic from low-level rendering commands, alongside a platform-agnostic abstraction layer that ensures consistent hardware interaction. Developers can further customize their workflow using a plugin-based API that allows for the injection of custom inspectors, tools, and asset importers directly into the editor interface. The platform supports high-performance simulation through a variant-based dynamic typing system and centralized resource management, which handles memory-efficient sharing of textures, models, and audio data. The engine also provides extensive developer tooling for compiling custom binaries and configuring build parameters to meet specific production requirements. Comprehensive documentation, including an offline-accessible class reference and community-maintained tutorials, is available to assist with project development and engine mastery.
C++game-developmentgame-enginegamedev - ggml-org/llama.cpp
ggml-org/llama.cpp
95,400Llama.cpp is an inference engine designed for the local execution of text-based and multimodal language models on consumer hardware. It provides a core environment for running models that process both text and image inputs, utilizing hardware-accelerated backends to optimize performance across diverse CPU and GPU architectures. The project distinguishes itself by offering a lightweight HTTP server that adheres to standard API specifications, enabling chat completion, embeddings, and reranking services. It includes a suite of tools for model quantization and conversion, which reduces memory usage and improves performance, alongside a command-line interface for managing chat templates and inference parameters. The ecosystem further supports structured data generation through grammar-based output constraints and provides diagnostic utilities for visualizing computational graphs. Comprehensive documentation is available, including a reference matrix that details the compatibility of computational operations across supported hardware backends.
C++ggml - home-assistant/core
home-assistant/core
84,936Home Assistant is a centralized home automation platform designed to orchestrate diverse internet-connected devices and services. It functions as a local-first control system that normalizes heterogeneous hardware protocols into a unified set of entities, attributes, and services. The core architecture relies on an event-driven state bus and a modular integration model, allowing the system to manage state changes and communicate across decoupled components through standardized interfaces. The platform distinguishes itself through a highly flexible, declarative configuration framework that allows users to define system behavior, automations, and entity settings using structured text files. It features a reactive automation engine that processes complex logic sequences triggered by state changes, temporal events, or external webhooks. To support advanced users, the system includes a template-based logic engine for dynamic data processing and a blueprint system that enables the reuse of pre-configured automation templates. Beyond basic orchestration, the project provides a comprehensive suite of administrative and diagnostic tools. This includes granular identity and access management, energy monitoring for various utilities, and sophisticated organizational features like area, floor, and label management. The system also offers extensive developer utilities, such as real-time state inspection, automation execution tracing, and live template debugging, to assist in maintaining and troubleshooting complex configurations. The system is configured primarily through YAML files, which are parsed and validated at runtime to ensure consistency across the integration ecosystem.
Pythonasynciohacktoberfesthome-automation - hacksider/Deep-Live-Cam
hacksider/Deep-Live-Cam
79,568Deep-Live-Cam is a generative video transformation tool designed for real-time facial manipulation and cinematic enhancement. It functions as a local-first AI runtime, performing all media processing directly on the user's hardware to ensure complete data privacy without external network dependencies. By utilizing a high-performance processing pipeline, the application enables live face swapping and interactive video modifications during active streaming sessions or on pre-recorded media. The system distinguishes itself through a hardware-abstraction execution layer that dynamically routes compute tasks to available graphics hardware, such as CUDA or CoreML backends. This architecture supports complex operations like multi-face mapping, where distinct target faces are applied to multiple subjects simultaneously, and preserves original mouth movements to maintain natural speech synchronization. To ensure visual fidelity, the engine employs precision mask-based blending and generative detail restoration, effectively integrating source features into target video geometry. Beyond core transformation capabilities, the application includes tools for cinematic rendering, such as real-time color grading and frame interpolation. It manages system resources through chunked memory and frame-based stream processing, which prevents crashes during intensive workloads and maintains stable performance. The interface is designed for focused workflows, offering distraction-free modes and automated projection window management to streamline the user experience during live operations.
Pythonaiai-deep-fakeai-face