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.

PyTorch training recipes for neural networks v1.0.0

Use this resource to diagnose and improve the training of a PyTorch neural network

New

  • 486 downloads
  • 247 likes
  • 120 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install ml-training-recipes-pytorch

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 helps you choose the architecture, optimizer and learning rate schedule, tune mixed precision, then diagnose an exploding loss or a memory error. It covers LLMs, vision, diffusion and other domains.

Mostly picked by

Updated November 2, 2026

Card for the “PyTorch training recipes for neural networks” skill

Details

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

Similar resources

View all

What you get

  • SKILL.md: training loop, optimizers, scheduling and debugging
  • references/optimizers.md, architecture.md, scaling-and-selection.md, experiment-loop.md, domain-specific.md and biomedical.md: optimizers, architectures, scaling laws and domains

Good fit for

  • Your loss diverges or your GPU runs out of memory
  • You choose an architecture or optimizer for a new project

Not for

    Tags

    • training
    • pytorch
    • debugging
    • optimizers

    About

    You say

    “My loss explodes after 2000 steps, help me diagnose it and tune the learning rate.”

    Card for the “PyTorch training recipes for neural networks” skill

    Skill published by dailycafi 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 diagnose and improve the training of a PyTorch neural network
    3. Review and adjust the result

    Content preview

    Files

    • LICENSE1.1 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/architecture.md10.3 KB
    • references/biomedical.md20.8 KB
    • references/domain-specific.md19.2 KB
    • references/experiment-loop.md4.5 KB
    • references/optimizers.md10.6 KB
    • references/scaling-and-selection.md16.9 KB
    • SKILL.md11.2 KB

    Origin and licence

    Listed unmodified

    Author
    dailycafi
    Source
    github.com/orchestra-research/ai-research-skills/tree/773a52944ba4747a18bd4ae9ade53fff041adcbc/10-optimization/ml-training-recipes
    Licence
    MIT

    Notice

    « ml-training-recipes » (dailycafi, 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.