midea-ai / Aidget
Ai edge toolbox，专门面向边端设备尤其是嵌入式RTOS平台，AI模型部署工具链，包括模型推理引擎和模型压缩工具
☆169Updated 11 months ago
Related projects ⓘ
Alternatives and complementary repositories for Aidget
- 【grps接入trtllm】通过GPRS+TensorRT-LLM+Tokenizers.cpp实现纯C++版高性能OpenAI LLM服务，支持chat和function call模式，支持ai agent，支持分布式多卡推理，支持多模态，支持gradio聊天界面。☆92Updated 2 weeks ago
- 【深度学习模型部署框架】支持tf/torch/trt/trtllm/vllm以及更多nn框架，支持dynamic batching、streaming模式，支持python/c++双语言，可限制，可拓展，高性能。帮助用户快速地将模型部署到线上，并通过http/rpc接口方式…☆165Updated last week
- 模型部署白皮书(CUDA|ONNX|TensorRT|C++)🚀🚀🚀☆186Updated 2 months ago
- A repo that uses TensorRT to deploy wll-trained models.Support RTDETR,YOLO-NAS,YOLOV5,YOLOV6,YOLOV7,YOLOV8,YOLOX.☆110Updated last year
- TengineGst is a streaming media analytics framework, based on GStreamer multimedia framework, for creating varied complex media analytics…☆74Updated 3 years ago
- 🚀 Do not need libtorch, pure C++ TensorRT deploys SOLOv2 etc, which can be quickly ported to NX/TX2.☆50Updated 2 years ago
- An ultra fast tiny model for lane detection, using onnx_parser, TensorRTAPI, torch2trt to accelerate. our model support for int8, dynamic…☆143Updated 3 years ago
- YOLOv5 pruning on COCO Dataset☆89Updated last year
- Algorithm acceleration landing framework, let you complete the development of algorithm at low cost.eg: Facedetect, FaceLandmark..☆90Updated 3 years ago
- 目标检测，采用yolov8作为基准模型，数据集采用VisDrone2019，带有自己的改进策略☆52Updated 4 months ago
- 使用deepspeed从头开始训练一个LLM,经过pretrain和sft阶段,验证llm学习知识、理解语言、回答问题的能力☆155Updated 4 months ago
- TVM Documentation in Chinese Simplified / TVM 中文文档☆957Updated this week
- SegmentAnything-OnnxRunner is an example using Meta AI Research's SAM onnx model in C++.The encoder and decoder of SAM are decoupled in t…☆100Updated 11 months ago
- An acceleration library that supports arbitrary bit-width combinatorial quantization operations☆226Updated last month
- Haohanyh-HAMOS-ProjectX Hi3861+华为云IoTDA+Android APP......☆31Updated 2 years ago
- ☆127Updated 10 months ago
- 从预训练到强化学习的中文llama2☆94Updated last year
- Detect known and unknown objects in the open world（具有区分已知与未知能力的全新检测器））☆91Updated last year
- DAN: Unfolding the Alternating Optimization for Blind Super Resolution☆72Updated 3 years ago
- ☆100Updated 2 years ago
- 系统基于 ncurses 图形库开发，提供多种用户身份验证功能，又针对不同身份的用户提供不同的设备控制和应用权限。项目使用 make 工具进行编译，可自动配置软链接，加载驱动模块，具体细节见仓库 README 文档。☆12Updated 2 years ago
- 该代码与B站上的视频 https://www.bilibili.com/video/BV18L41197Uz/?spm_id_from=333.788&vd_source=eefa4b6e337f16d87d87c2c357db8ca7 相关联。☆60Updated last year
- This repo is the official implementation of "BEVTrack: A Simple and Strong Baseline for 3D Single Object Tracking in Bird's-Eye View".☆76Updated 3 weeks ago
- Chat-Style-Bot是一个聊天风格模仿大语言模型，通过分析和学习微信聊天记录，可模仿你的说话风格（口头禅等），并可接入微信和你的朋友们自动聊天。Chat-Style-Bot is a chat style imitating llm. By analyzing an…☆84Updated 3 months ago
- A high-performance web server based on the multiplexer actor mode, with an underlying implementation using the epoll model☆46Updated 6 months ago
- An onnx-based quantitation tool.☆70Updated 10 months ago
- 动手构建一个完整的神经网络; Hands-on construction of a complete neural network☆13Updated last year
- MIXQ: Taming Dynamic Outliers in Mixed-Precision Quantization by Online Prediction☆82Updated 3 weeks ago