Every premium and members-only resource for $9.99/month, as a no-commitment subscription. See the plans

PalateKit

Sign in
Menu

Request a resource

Describe the resource you are missing in one sentence. We will email you when it is published.

0 / 280

A page, a video or an image online.

Unlock PalateKit

Unlock everything

Plan

Paying means accepting our terms of sale.

Secure payment by Stripe. Receipts and invoices sent by email.

Welcome back

Sign in to install, download or save this resource. A free account is enough for free resources.

ML training health checks v1.0.0

Use this resource to monitor a model training run and catch broken runs early

New

  • 460 downloads
  • 188 likes
  • 173 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install ml-training-health-checks

With no option, the skill goes to ~/.agents/skills/; add --agent claude-code (or codex, cursor…) to target an agent.

First time: connect the CLI with

npx --yes palatekit@latest login

The agent periodically reads WandB metrics while training runs and judges whether something is wrong: NaN values, loss divergence or an idle GPU. It acts on the finding and asks a second reviewer to settle doubtful cases. You avoid paying for GPU hours spent on a broken run.

Mostly picked by

Updated November 2, 2026

Card for the “ML training health checks” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
MIT

Similar resources

View all

What you get

  • SKILL.md: metric reading, judgment and actions
  • SKILL.md: CronCreate scheduling example and usage rules

Good fit for

  • A GPU training run is underway and you want to avoid wasted hours
  • You track runs in WandB and want automated health checks

Not for

    Tags

    • training
    • wandb
    • gpu
    • monitoring

    About

    You say

    “Watch the current run and tell me if the loss diverges or a GPU sits idle.”

    Card for the “ML training health checks” skill

    Skill published by wanshuiyin under the MIT license, delivered in its original version.

    Show more

    How to use it

    1. Install the skill in your agent's skills folder
    2. Ask your agent to monitor a model training run and catch broken runs early
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • SKILL.md6.7 KB

    Origin and licence

    Listed unmodified

    Author
    wanshuiyin
    Source
    github.com/wanshuiyin/auto-claude-code-research-in-sleep/tree/18bceb3c71c638bef58096274e1c9a6d323dbc0a/skills/training-check
    Licence
    MIT

    Notice

    « training-check » (wanshuiyin, MIT) livré dans sa version d'origine : fiche en français et en anglais par PalateKit.

    User reviews

    –

    No ratings yet

    Loading reviews…

    Delete your review?

    Your rating and comment will be removed from the page.