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.

Train an OCDNet scene text detector v1.0.0

Use this resource to train and evaluate an arbitrarily oriented text detector on natural images

New

  • 499 downloads
  • 183 likes
  • 240 saves

Free

Install with the CLI

In a terminal, from the folder you want:

npx --yes palatekit@latest install scene-text-detector-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 writes the spec from your data, runs the TAO action you need (training, evaluation, pruning, export or inference) and checks the results. It also covers the TensorRT steps and common errors.

Mostly picked by

Updated November 2, 2026

Card for the “Train an OCDNet scene text detector” skill

Details

Generates with
Your agent's model
Licence
Apache-2.0

Similar resources

View all

What you get

  • SKILL.md : actions, requirements and common errors
  • references/spec_template_export.yaml : export spec template
  • references/spec_template_prune.yaml : pruning spec template

Good fit for

  • You want to read signs or panels in photos
  • You need to prune or export a text detector

Not for

    Tags

    • ocdnet
    • text detection
    • tao
    • training
    • ocr

    About

    You say

    “Train OCDNet on my street photos with angled text.”

    Card for the “Train an OCDNet scene text detector” 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 evaluate an arbitrarily oriented text detector on natural images
    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.yaml6.3 KB
    • references/spec_template_deploy_evaluate.yaml0.8 KB
    • references/spec_template_deploy_gen_trt_engine.yaml0.5 KB
    • references/spec_template_deploy_inference.yaml0.4 KB
    • references/spec_template_evaluate.yaml3.5 KB
    • references/spec_template_export.yaml3.2 KB
    • references/spec_template_gen_trt_engine.yaml3.5 KB
    • references/spec_template_inference.yaml3.5 KB
    • references/spec_template_prune.yaml3.2 KB
    • references/spec_template_quantize.yaml3.1 KB
    • references/spec_template_retrain.yaml3.1 KB
    • references/spec_template_train.yaml3.1 KB
    • references/tao-deploy-ocdnet.md6.3 KB
    • references/tao-deploy-ocdnet.skill_info.yaml3.4 KB
    • schemas/evaluate.schema.json62.7 KB
    • schemas/export.schema.json57.6 KB
    • schemas/gen_trt_engine.schema.json64.7 KB
    • schemas/inference.schema.json62.7 KB
    • schemas/manifest.json27.1 KB
    • schemas/prune.schema.json57.2 KB
    • schemas/quantize.schema.json55.1 KB
    • schemas/retrain.schema.json55.1 KB
    • schemas/train.schema.json55.1 KB
    • skill-card.md4.5 KB
    • SKILL.md14.5 KB
    • skill.oms.sig10.3 KB

    Origin and licence

    Listed unmodified

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

    Notice

    « tao-train-ocdnet » (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.