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.

Training a TAO masked autoencoder v1.0.0

Use this resource to pretrain and fine-tune a TAO MAE backbone with self-supervised learning

New

  • 352 downloads
  • 202 likes
  • 158 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install masked-autoencoder-pretraining

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 sets up the pretrain and finetune stages, fills train.pretrained_model_path for MAE weights when fine-tuning, then guides evaluation, export and inference. For TensorRT actions it points to the skill's deployment guide first. You get a visual backbone suited to your dataset.

Mostly picked by

Updated November 2, 2026

Card for the “Training a TAO masked autoencoder” skill

Details

Works in
Install guides
Generates with
Your agent's model
Licence
Apache-2.0

Similar resources

View all

What you get

  • references/spec_template_train.yaml: training spec
  • references/spec_template_export.yaml: export spec
  • references/tao-deploy-mask-auto-encoder.md: TensorRT deployment guide

Good fit for

  • You want a visual backbone pretrained on your own images
  • You fine-tune MAE weights for a vision task

Not for

    Tags

    • MAE
    • self-supervised learning
    • vision
    • pretraining

    About

    You say

    “Pretrain a MAE on my unlabeled images, then fine-tune it for my classification task.”

    Card for the “Training a TAO masked autoencoder” skill

    Skill published by nvidia under the Apache-2.0 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 pretrain and fine-tune a TAO MAE backbone with self-supervised learning
    3. Review and adjust the result

    Content preview

    Files

    • BENCHMARK.md7.2 KB
    • config/skillspector-baseline.yaml0.8 KB
    • evals/evals.json0.8 KB
    • LICENSE11.3 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/skill_info.yaml3.4 KB
    • references/spec_template_deploy_gen_trt_engine.yaml0.8 KB
    • references/spec_template_evaluate.yaml1.8 KB
    • references/spec_template_export.yaml1.9 KB
    • references/spec_template_gen_trt_engine.yaml2 KB
    • references/spec_template_inference.yaml1.8 KB
    • references/spec_template_train.yaml1.7 KB
    • references/tao-deploy-mask-auto-encoder.md3.5 KB
    • references/tao-deploy-mask-auto-encoder.skill_info.yaml1.2 KB
    • schemas/evaluate.schema.json30.1 KB
    • schemas/export.schema.json31 KB
    • schemas/gen_trt_engine.schema.json32.9 KB
    • schemas/inference.schema.json30.1 KB
    • schemas/manifest.json8 KB
    • schemas/train.schema.json27.5 KB
    • skill-card.md4.5 KB
    • SKILL.md11.4 KB
    • skill.oms.sig8.4 KB

    Origin and licence

    Listed unmodified

    Author
    nvidia
    Source
    github.com/nvidia/skills/tree/ca116d60beafe1b4459f0ab790b7b450912c56b0/skills/tao-train-mask-auto-encoder
    Licence
    Apache-2.0

    Notice

    « tao-train-mask-auto-encoder » (nvidia, Apache-2.0) 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.