#python #bytetrack #multi_object_tracking #oc_sort #sort

Trackers is a simple Python library (pip install trackers) for multi-object tracking that plugs into any detection model like YOLO. Use it via CLI on videos/webcams or in Python code with trackers like ByteTrack (top performer on MOT17/SportsMOT benchmarks) to add labels and trajectories. Evaluate with MOT metrics too. Benefit: Quickly add reliable object tracking to your computer vision projects for real-time apps like traffic or sports analysis, saving time on custom code.

https://github.com/roboflow/trackers

#python

Hugging Face Skills are ready-to-use folders with instructions, scripts, and tools for AI agents to handle tasks like creating datasets, training models, running evaluations, managing jobs, and publishing papers. They work seamlessly with tools like Claude Code, OpenAI Codex, Gemini CLI, and Cursor—just install via simple commands and mention the skill in your instructions, such as "Use the HF model trainer skill." This saves you time by automating complex Hugging Face Hub operations, letting your agent execute them accurately without manual coding.

https://github.com/huggingface/skills

#python #agents #claude #cursor #databricks #vibecoding

The Databricks AI Dev Kit enhances AI-driven development by providing your coding assistant (Claude Code, Cursor, etc.) with trusted Databricks knowledge and best practices. It includes a Python library, MCP server with 50+ tools, markdown skills teaching Databricks patterns, and a web-based builder app. You can build Spark pipelines, jobs, dashboards, knowledge assistants, and deploy ML models faster and smarter. The benefit is that your AI coding assistant gains direct access to Databricks functionality and patterns, enabling you to develop data and AI applications more efficiently with built-in governance and best practices.

https://github.com/databricks-solutions/ai-dev-kit