Miss. Wu's repositories
align-anything
Align Anything: Training All-modality Model with Feedback
AML-Monitoring-Engine
AML end to end system
Awesome-GraphRAG
A curated list of resources on graph-based retrieval-augmented generation (GraphRAG) for customized large language models.
Awesome-Object-Insertion
A curated list of papers, code and resources pertaining to image composition/compositing or object insertion/addition/compositing, which aims to generate realistic composite image.
BOAZ_beta
Multilayered AV/EDR Evasion Framework
custom-lens-wa-hub
Provide JSON file template that demonstrate how to create customize Well-Architected reviews using Custom lenses.
Data-Labeling
数据标注是一款专门对文本数据进行处理和标注的工具，通过简化快捷的文本标注流程和动态的算法反馈，支持用户快速标注关键词并能通过算法持续减少人工标注的成本和时间。数据标注的过程先由人工标注构筑基础，再由自动标注反哺人工标注，最后由人工标注进行纠偏，从而大幅度提高标注的精准度和高效性。数据标注是一个完全开源的项目，无商业版，但是需要依赖开源的数字底座进行人员岗位管控。各类词库结果会定期在本平台公开。
DataFlow-Engine
数据流引擎是一款面向数据集成、数据同步、数据交换、数据共享、任务配置、任务调度的底层数据驱动引擎。数据流引擎采用管执分离、多流层、插件库等体系应对大规模数据任务、数据高频上报、数据高频采集、异构数据兼容的实际数据问题。
eiam
以开源为核心的IDaas/IAM平台，用于管理企业内员工账号、权限、身份认证、应用访问，帮助整合部署在本地或云端的内部办公系统、业务系统及三方 SaaS 系统的所有身份，实现一个账号打通所有应用的服务。
电子邮件是一款简化的具备邮件服务器的企业邮箱，支持在将其他主流邮箱的邮件进行导入后自主控制邮件数据安全。电子邮件具备较为简洁的界面风格，以其简洁精确的功能和小巧安全的架构便于企业和政府根据业务要求进行二次开发。电子邮件需要依赖开源的数字底座进行人员岗位管控。
evogp
A GPU-accelerated library for Tree-based Genetic Programming, leveraging PyTorch and custom CUDA kernels for high-performance evolutionary computation. It supports symbolic regression, classification, and policy optimization with advanced features like multi-output trees and benchmark tools.
fit-framework
FIT: 企业级AI开发框架，提供多语言函数引擎（FIT）、流式编排引擎（WaterFlow）及Java生态的LangChain替代方案（FEL）。原生/Spring双模运行，支持插件热插拔与智能聚散部署，无缝统一大模型与业务系统。
GENERator
GENERator: A Long-Context Generative Genomic Foundation Model
hallo3
Hallo3: Highly Dynamic and Realistic Portrait Image Animation with Diffusion Transformer Networks
manuscript-core
Manuscript is a revolutionary blockchain data streaming framework. With Manuscript, you can seamlessly integrate on-chain and off-chain data into target data storage for unrestricted querying and analysis
MARS
The official implementation of MARS: Unleashing the Power of Variance Reduction for Training Large Models
mini-nanoGPT
One-click training of your own GPT. Training a GPT has never been easier. / 训练一个GPT原来可以这么简单？
Network-Drive
网络硬盘是通过存储、分类、检索、分享、协作、下发、回收、展示等方式管理文档、文件、图片、音频、视频等资料的工具。网络硬盘擅长在国产的私有化环境中管控文档权限、存储空间分配、安全加密、链接分享，同时支持一定轻量级的文件任务收发。网络硬盘需要依赖开源的数字底座进行人员岗位管控。
PayGuard
A machine learning-driven solution designed to detect fraudulent activities in bank payment systems
PyramidKV
The Official Implementation of PyramidKV: Dynamic KV Cache Compression based on Pyramidal Information Funneling
react-native-turbo-image
Performant image component for React Native
Sa2VA
🔥 Sa2VA: Marrying SAM2 with LLaVA for Dense Grounded Understanding of Images and Videos
Sheas-Cealer
Just Ceal It (可用于无代理合法抵御网络监听和开展网络研究)
TransProPy
A python package that integrate algorithms and various machine learning approaches to extract features (genes) effective for classification and attribute them accordingly.
uco3d
Uncommon Objects in 3D dataset
xDiT
xDiT: A Scalable Inference Engine for Diffusion Transformers (DiTs) with Massive Parallelism
ZhiLight
A highly optimized LLM inference acceleration engine for Llama and its variants.