tensorrt

#计算机科学#NVIDIA®TensorRT™是一款用于在NVIDIA GPU上进行高性能深度学习推理的SDK。此存储库包含TensorRT的开源组件。

YOLOX is a high-performance anchor-free YOLO, exceeding yolov3~v5 with MegEngine, ONNX, TensorRT, ncnn, and OpenVINO supported. Documentation: https://yolox.readthedocs.io/

翻译 - YOLOX是一款高性能的无锚YOLO，超越yolov3~v5，支持ONNX、TensorRT、ncnn、O​​penVINO。

#计算机科学#The user analytics platform for LLMs

翻译 - Easy-to-use library to boost AI inference leveraging multiple DL compilers

Implementation of popular deep learning networks with TensorRT network definition API

翻译 - 使用TensorRT网络定义API实施流行的深度学习网络

#计算机科学#BoxMOT: pluggable SOTA tracking modules for segmentation, object detection and pose estimation models

翻译 - 使用 YOLO v5 和深度排序的实时多目标跟踪器

An easy to use PyTorch to TensorRT converter

翻译 - 易于使用的PyTorch到TensorRT转换器

PyTorch ,ONNX and TensorRT implementation of YOLOv4

翻译 - YOLOv4的PyTorch，ONNX和TensorRT实现

#计算机科学#由腾讯优图实验室开源的高性能、轻量级神经网络推理框架，同时拥有跨平台、高性能、模型压缩、代码裁剪等众多突出优势。

#计算机科学#DAMO-YOLO: a fast and accurate object detection method with some new techs, including NAS backbones, efficient RepGFPN, ZeroHead, AlignedOTA, and distillation enhancement.

🔥🔥🔥🔥 (Earlier YOLOv7 not official one) YOLO with Transformers and Instance Segmentation, with TensorRT acceleration! 🔥🔥🔥

翻译 - 🔥🔥🔥🔥 带有 Transformers 和 Instance Segmentation 的 YOLO，带有 TensorRT 加速！ 🔥🔥🔥

#安卓#⚡️An Easy-to-use and Fast Deep Learning Model Deployment Toolkit for ☁️Cloud 📱Mobile and 📹Edge. Including Image, Video, Text and Audio 20+ main stream scenarios and 150+ SOTA models with end-to-end ...

OpenMMLab Model Deployment Framework

翻译 - OpenMMLab 模型部署框架

#计算机科学#PyTorch/TorchScript/FX compiler for NVIDIA GPUs using TensorRT

翻译 - 面向NVIDIA GPU的PyTorch JIT的提前编译

#大语言模型#Generative AI reference workflows optimized for accelerated infrastructure and microservice architecture.

🍅🍅🍅YOLOv5-Lite: Evolved from yolov5 and the size of model is only 900+kb (int8) and 1.7M (fp16). Reach 15 FPS on the Raspberry Pi 4B~

翻译 - shufflev2-yolov5：更轻、更快、更容易部署。由yolov5进化而来，模型大小只有1.7M（int8）和3.3M（fp16）。当输入大小为320×320时，它可以在树莓派4B上达到10+ FPS~

#安卓#YOLOv4, YOLOv4-tiny, YOLOv3, YOLOv3-tiny Implemented in Tensorflow 2.0, Android. Convert YOLO v4 .weights tensorflow, tensorrt and tflite

翻译 - YOLOv4，YOLOv4-tiny，YOLOv3，YOLOv3-tiny在Tensorflow 2.0，Android中实现。转换YOLO v4 .weights张量流，张量和tflite

A nearly-live implementation of OpenAI's Whisper.

YOLO5Face: Why Reinventing a Face Detector (https://arxiv.org/abs/2105.12931) ECCV Workshops 2022)

TensorRT MODNet, YOLOv4, YOLOv3, SSD, MTCNN, and GoogLeNet

翻译 - TensorRT YOLOv4，YOLOv3，SSD，MTCNN和GoogLeNet

Efficient, scalable and enterprise-grade CPU/GPU inference server for 🤗 Hugging Face transformer models 🚀

翻译 - 在 Nvidia Triton 服务器上部署优化的基于变压器的模型