huggingface-trackio

Log and visualize ML training metrics with trackio and optional TRL integration.

Updated Aug 23, 2026

Search Tags:#metrics #dashboard #alerts #ml #space #trackio #experiment-tracking

Ready to run

npx skills add https://github.com/domattioli/DomI --skill huggingface-trackio-domattioli

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: huggingface-trackio
Source: https://github.com/domattioli/DomI/tree/main/plugins/huggingface-skills/skills/huggingface-trackio
Command: npx skills add https://github.com/domattioli/DomI --skill huggingface-trackio-domattioli

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill includes references (resource) components.

What problem does it solve?

Trackio enables seamless tracking of ML training runs by logging metrics, triggering alerts, and visualizing results in real-time dashboards, with optional HF Space synchronization and JSON output for automation.

Core Features & Use Cases

Logging metrics during training via Python API (trackio.log)
Firing alerts for diagnostics with trackio.alert()
Retrieving metrics and summaries via CLI commands
Remote dashboards via HF Space and JSON export for automation
TRL integration support for automatic metric logging with Transformers/TRL

Quick Start

Initialize a run with trackio.init(...), log metrics with trackio.log(...), and finalize with trackio.finish()

Frequently Asked Questions about huggingface-trackio

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I track ML experiment metrics in real-time during model training?▼

To track ML experiment metrics, initialize a run with trackio.init, log metrics with trackio.log, and finalize using trackio.finish to enable real-time dashboard monitoring.

Can I use trackio with TRL Transformers for automatic metric logging?▼

Yes, trackio provides TRL integration support, allowing you to automatically log training metrics and monitor model development workflows within the Transformers trainer.

What is the best way to set up alerts for ML training diagnostics?▼

The best way to set up alerts for ML training diagnostics is using the trackio.alert() function, which fires alerts directly during your model development workflow when specific conditions are met.

How do I retrieve logged experiment metrics from the command line?▼

You can retrieve logged experiment metrics and summaries from the command line by using trackio CLI commands, which query your training run data and return outputs for automation.

Does trackio support syncing dashboards to HF Spaces for persistent visibility?▼

Yes, trackio supports syncing dashboards to HF Spaces, enabling persistent remote visibility of your ML training metrics and real-time monitoring data across sessions.