burn-tracing-backend

Community

Visualize and analyze tensor operation performance in ML workflows.

AuthorAdrianEddy
Version1.0.0
Installs0

System Documentation

What problem does it solve?

This Skill provides detailed tracing and visualization of tensor operations, helping developers identify performance bottlenecks and fusion efficiency issues in machine learning models.

Core Features & Use Cases

  • Operation profiling: Records CPU timing, fusion events, and synchronization points during tensor computations.
  • Interactive visualizations: Generates HTML-based timelines, memory usage charts, and operation lists for performance analysis.
  • Use Case: Optimize a deep learning training pipeline by pinpointing slow fusion kernels and excessive syncs, then refining the model code accordingly.

Quick Start

Use the burn-tracing backend to trace a model inference and generate an HTML report for performance review.

Dependency Matrix

Required Modules

burn-tracing-backend

Components

scriptsreferencesassets

💻 Claude Code Installation

Recommended: Let Claude install automatically. Simply copy and paste the text below to Claude Code.

Please help me install this Skill:
Name: burn-tracing-backend
Download link: https://github.com/AdrianEddy/burn-tracing-backend/archive/main.zip#burn-tracing-backend

Please download this .zip file, extract it, and install it in the .claude/skills/ directory.
View Source Repository

Agent Skills Search Helper

Install a tiny helper to your Agent, search and equip skill from 471,000+ vetted skills library on demand.