Project Library

Innovative responsive Python notebook. This project is a responsive notebook designed specifically for Python, which automatically executes and updates the dependent code cells when interacting with the UI, ensuring consistency between the code and output. It is stored in pure Python files, making it easy to manage and run, and supports execution as a script or deployment as an interactive web application.

Chat with your SQL database. This project uses LLM+RAG+database technology to allow users to query SQL databases through natural language and generate SQL answers to your questions.

Ultimate camera streaming application with support for RTSP, RTMP, HTTP-FLV, WebRTC, MSE, HLS, MP4, MJPEG, HomeKit, FFmpeg, etc.

Python's data stream orchestration platform. If the programs for acquiring, cleaning, and processing data are considered as individual tasks, this project can integrate these tasks into a workflow, enabling their deployment, scheduling, and monitoring on a web platform.

A command-line tool to improve the readability of Git diffs. This is a command-line tool written in Rust that can display Git's diff, blame, and grep commands in a clearer and more beautiful way. It supports features such as displaying line numbers, syntax highlighting, and smarter inline difference displays, seamlessly integrating into the Git workflow.

Resolve production issues quickly. An open source observability platform unifying session replays, logs, metrics, traces, and errors.

Sample code and notebooks for Generative AI on Google Cloud

✨ Magical shell history

Convert any music library into a music production sample-library with ML

Generate a Twitter bio in seconds using AI and current bio

Sketch is an AI code-writing assistant for pandas users that requires no setup.

An open-source LLM-based research assistant.

The source ERP/Crm project is based on the advanced Mern stack (Node.js / Express.js / MongoDb / React.js) with Ant design (AntD) and Redux. It provides management functions such as invoices, inventory, accounting, and human resources, aiming to provide a comprehensive management solution for enterprises.

This project is a Pytorch version of semantic segmentation and scene parsing on MIT ADE20K dataset. It provides an open-source tool for semantic segmentation and scene parsing tasks, especially suitable for research and application in the field of computer vision.

A powerful open source TTS engine, supporting both Chinese and English, with more than 2000 different voices, as well as a special emotional synthesis function, supporting the synthesis of speech with a wide range of emotions such as happiness, excitement, sadness, anger, etc.