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

PalateKit

Menu

Request a resource

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

0 / 280

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.

Grounding DINO training for open-vocabulary detection v1.0.0

Use this resource to train and deploy a Grounding DINO model for text-guided detection with TAO

New

  • 465 downloads
  • 213 likes
  • 189 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install grounding-dino-open-vocabulary-training

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 prepares the training and deploy specs, distinguishing full weights from backbone-only initialization. You get configuration files for detection guided by a text prompt.

Mostly picked by

Updated November 2, 2026

Card for the “Grounding DINO training for open-vocabulary detection” skill

Details

Generates with
Your agent's model
Licence
Apache-2.0

Similar resources

View all

What you get

  • references/spec_template_export.yaml: export spec
  • references/spec_template_evaluate.yaml: evaluation spec
  • references/skill_info.yaml: container and action information

Good fit for

  • You want to detect objects described in natural language without relabeling
  • You have a full model to fine-tune rather than a backbone only

Not for

    Tags

    • grounding dino
    • open vocabulary
    • text
    • tao

    About

    You say

    “Prepare the export of a Grounding DINO model that detects hard hats from a text description.”

    Card for the “Grounding DINO training for open-vocabulary detection” 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 train and deploy a Grounding DINO model for text-guided detection with TAO
    3. Review and adjust the result

    Content preview

    Files

    • BENCHMARK.md7.1 KB
    • config/skillspector-baseline.yaml0.8 KB
    • evals/evals.json0.8 KB
    • LICENSE11.3 KB
    • NOTICE-PALATEKIT.md0.5 KB
    • references/skill_info.yaml6.2 KB
    • references/spec_template_deploy_evaluate.yaml0.4 KB
    • references/spec_template_deploy_gen_trt_engine.yaml0.4 KB
    • references/spec_template_deploy_inference.yaml0.4 KB
    • references/spec_template_evaluate.yaml3.4 KB
    • references/spec_template_export.yaml3.5 KB
    • references/spec_template_gen_trt_engine.yaml3.6 KB
    • references/spec_template_inference.yaml3.5 KB
    • references/spec_template_quantize.yaml3.3 KB
    • references/spec_template_train.yaml3.3 KB
    • references/tao-deploy-grounding-dino.md5.5 KB
    • references/tao-deploy-grounding-dino.skill_info.yaml3 KB
    • schemas/evaluate.schema.json58 KB
    • schemas/export.schema.json58.6 KB
    • schemas/gen_trt_engine.schema.json60.4 KB
    • schemas/inference.schema.json59 KB
    • schemas/manifest.json16.5 KB
    • schemas/quantize.schema.json54.6 KB
    • schemas/train.schema.json54.6 KB
    • skill-card.md4.3 KB
    • SKILL.md15.4 KB
    • skill.oms.sig9.4 KB

    Origin and licence

    Listed unmodified

    Author
    nvidia
    Source
    github.com/nvidia/skills/tree/ca116d60beafe1b4459f0ab790b7b450912c56b0/skills/tao-train-grounding-dino
    Licence
    Apache-2.0

    Notice

    « tao-train-grounding-dino » (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.