A public account focused on deep learning, computer vision, and autonomous driving perception algorithms, covering visual CV, neural networks, pattern recognition, related hardware and software configurations, and open-source projects.
This guide shows how to configure Claude Code to run on local or cloud models within five minutes, covering hardware requirements, recommended models, step‑by‑step installation for Ollama, llama.cpp, LM Studio, and cloud‑based options, plus performance and cost comparisons.
This article explains what Agent Skills are, why a curated skill library is valuable, and reviews eight actively maintained GitHub repositories—detailing their structure, core capabilities, integration points, and practical usage examples for building production‑grade AI agents.
Claw Code, an open‑source Python‑and‑Rust reimplementation of Anthropic’s Claude Code agent, exploded to over 100 k stars within hours after a leaked .map file revealed 510 k lines of the original TypeScript, and the article dissects its creator, architecture, features, and legal gray area.
Claude Code, Anthropic's CLI AI programming assistant, offers six extensible mechanisms—Skills, MCP, Sub‑agents, Agent Teams, Hooks, and Plugins—each explained with purpose, setup steps, concrete examples, and practical guidance on when and how to combine them for robust AI‑driven workflows.
This guide explains the concept of LLM "Skills", shows how to organize skill directories for Claude and Copilot, walks through creating a "prepare‑pr" skill with a SKILL.md file, integrates Bash scripts for git checks, and demonstrates testing and extending the skill with additional checks and templates.
This step‑by‑step tutorial shows how to install NanoBot, configure it with a DeepSeek API key, create a YOLO‑based pedestrian detection skill via natural‑language commands, test the generated code, and extend the output to JSON, demonstrating AI agents in Python.
The article surveys the emerging personal AI‑assistant ecosystem—including OpenClaw, NanoBot, PicoClaw, IronClaw, and ZeroClaw—detailing each project's origins, technology stack, performance metrics, and design goals, then dives deep into OpenClaw's layered memory, six‑stage execution pipeline, tool‑skill framework, and five core architectural principles.
The article explains how manufacturers can shift from defect‑specific vision models to anomaly detection by leveraging pre‑trained object‑detection networks, visualising feature maps, and applying memory‑bank methods such as PaDiM and PatchCore, with the open‑source Anomalib library as a ready‑to‑use solution.
This article explains how to train a simple fully‑connected autoencoder on defect‑free images, use reconstruction error to highlight anomalies in industrial parts, and convert the error into a single metric that cleanly separates good from defective components.
The article explains how contrastive learning replaces fixed‑category visual training with a relationship‑based approach, detailing the dual‑encoder architecture, cosine similarity loss, batch scaling, temperature control, zero‑shot capabilities, scalability from web data, and the method's strengths and limitations in modern multimodal AI.
