6 रिपॉजिटरी
Systems that optimize and execute mathematical operations by constructing directed acyclic graphs of tensors and operators.
Explore 6 awesome GitHub repositories matching scientific & mathematical computing · Graph-Based Execution Engines. Refine with filters or upvote what's useful.
This repository serves as a centralized collection of state-of-the-art deep learning architectures and reference implementations designed for research and application development. It provides a comprehensive toolkit for computer vision and natural language processing, offering pre-built models and training pipelines for tasks ranging from image classification and object detection to complex sequence modeling. The project distinguishes itself by providing a flexible execution harness that manages the entire training lifecycle, including data ingestion and backpropagation. It supports scalable
Constructs directed acyclic graphs of tensors and operators to enable high-performance execution of mathematical operations.
This project is a cross-platform machine learning inference engine designed to execute pre-trained models across diverse operating systems and hardware environments. It functions as a standardized execution framework that manages the entire lifecycle of model inference, from loading and graph optimization to hardware-accelerated execution and generative sequence management. The runtime distinguishes itself through a highly modular architecture that decouples model logic from hardware-specific kernels. By utilizing an execution provider abstraction, it enables developers to offload computation
Optimizes and executes machine learning models by representing them as directed acyclic graphs of tensors and operators.
HVM2 is a high-performance execution environment for pure functional programs, implemented as a systems-level runtime in Rust. It functions as a massively parallel functional runtime that uses interaction combinators to achieve automatic parallelism across multi-core CPUs and GPUs. The project distinguishes itself by using a graph-rewriting computational model to execute programs via local reduction rules, which eliminates the need for manual locks or atomic operations. It employs beta-optimal reduction and lazy evaluation to optimize higher-order functions and eliminate redundant computation
Uses a graph-based execution engine to distribute functional programs across multiple cores without locks.
Cactus is an on-device AI inference engine designed for executing large language models, vision models, and speech-to-text systems on mobile and wearable hardware. It provides a programmable tensor computation graph for defining sequences of matrix operations and activation functions, alongside a local retrieval augmented generation framework that grounds model responses using local text files. The project features a multiplatform SDK with language bindings for integrating AI capabilities into mobile applications and a model conversion system that transforms external model formats for optimiz
Executes mathematical workflows as a sequence of tensor operations and activation functions via directed acyclic graphs.
Neon is a deep learning framework and hardware-abstraction machine learning stack used for designing, training, and deploying neural network architectures. It functions as a graph-based computation engine that utilizes just-in-time kernel compilation to optimize machine code for tensors. The platform decouples model definitions from execution kernels, allowing it to support multiple CPU and GPU backends. This architecture enables the distribution of computational workloads across parallelized hardware environments to increase processing speed and overall efficiency. The system covers the ful
Implements an execution engine that optimizes mathematical operations by constructing directed acyclic graphs of tensors.
यह प्रोजेक्ट डीप लर्निंग अवधारणाओं और न्यूरल नेटवर्क आर्किटेक्चर को सिखाने के लिए डिज़ाइन की गई इंटरैक्टिव नोटबुक के एक स्ट्रक्चर्ड पाठ्यक्रम से युक्त एक शैक्षिक संसाधन है। यह TensorFlow 2 फ्रेमवर्क और Keras API के साथ व्यावहारिक अनुभव प्रदान करने पर केंद्रित है, जो उपयोगकर्ताओं को मशीन लर्निंग तकनीकों में महारत हासिल करने के लिए व्यावहारिक अभ्यासों के माध्यम से मार्गदर्शन करता है। रिपॉजिटरी उच्च-प्रदर्शन कंप्यूटिंग के लिए तकनीकी आवश्यकताओं के साथ निर्देशात्मक सामग्री को जोड़कर खुद को अलग करती है। इसमें ग्राफिक्स प्रोसेसिंग यूनिट्स पर हार्डवेयर-एक्सेलेरेटेड प्रशिक्षण का समर्थन करने के लिए स्थानीय डेवलपमेंट वातावरण को कॉन्फ़िगर करने के लिए विशिष्ट गाइड शामिल हैं, यह सुनिश्चित करते हुए कि शिक्षार्थी अपने मॉडल के लिए गहन गणितीय गणना कर सकें। प्लेटफॉर्म न्यूरल नेटवर्क आर्किटेक्चर के रैपिड प्रोटोटाइपिंग से लेकर डेटा साइंस प्रयोगों के निष्पादन तक, डेवलपमेंट वर्कफ़्लो की एक श्रृंखला का समर्थन करता है। यह दस्तावेज़ों के भीतर गणितीय समीकरणों और कोड को रेंडर करने के लिए उपयोगिताएं प्रदान करती है, और यह स्थानीय ऑफ़लाइन डेवलपमेंट और क्लाउड-आधारित निष्पादन वातावरण दोनों को समायोजित करती है।
Defines neural network operations as a series of mathematical nodes that are optimized and executed by a high-performance backend engine.