tao-train-dino
installsforks
How to install tao-train-dino?
nvidia/skills/tao-train-dino
$npx -y skills add nvidia/skills --skill tao-train-dinoDINO (DETR with Improved DeNoising Anchor Boxes) for 2D object detection. Transformer-based detector with
SKILL.md
| 1 | # DINO |
| 2 | |
| 3 | DINO (DETR with Improved DeNoising Anchor Boxes) for 2D object detection. Transformer-based detector with denoising training, multi-scale features, and optional distillation support. |
| 4 | |
| 5 | Uses pretrained backbone weights (e.g. ResNet-50 ImageNet). Set `model.pretrained_backbone_path` for backbone-only or `train.pretrained_model_path` for full model. |
| 6 | |
| 7 | ## When To Use |
| 8 | |
| 9 | Train, evaluate, export, distill, quantize, or run inference for a TAO DINO 2D object detector. |
| 10 | |
| 11 | For TAO Deploy TensorRT actions (`gen_trt_engine`, TensorRT `evaluate`, and |
| 12 | TensorRT `inference`), read `references/tao-deploy-dino.md` first. Deploy spec templates live |
| 13 | in this skill's `references/` folder with the `spec_template_deploy_*.yaml` |
| 14 | prefix. |
| 15 | |
| 16 | ## Reference Map |
| 17 | |
| 18 | - `references/dino-data-specs.md` — dataset contracts, per-action dataset requirements, per-action spec-override examples (train, evaluate, export, deploy/gen_trt_engine, inference, quantize, distill), data-source arrays, checkpoint inference, and dataset layout. |
| 19 | - `references/dino-actions-errors.md` — important parameters, default values, evaluate/export defaults, hardware, and the full error-pattern catalog. |
| 20 | - `references/dino-tuning-multigpu.md` — full AutoML/HPO notes (metrics, hyperparameters, extractor) and multi-GPU spec consistency. |
| 21 | - `references/dino-automl-sdk.md` — AutoML metrics, SDK orchestration internals, data-source gap, and spec-param/parent-model inference. |
| 22 | - `references/tao-deploy-dino.md` — TensorRT deploy workflow. |
| 23 | - `references/detailed-guide.md` — map to the detailed model guide. |
| 24 | |
| 25 | ## Dataclass Schemas |
| 26 | |
| 27 | Generated TAO Core schemas are packaged in `schemas/<action>.schema.json`, with `schemas/manifest.json` listing available actions. Each generated schema also emits `references/spec_template_<action>.yaml` from the schema top-level `default` field. AutoML enablement is declared at the model layer in `references/skill_info.yaml` via `automl_enabled`. Runnable AutoML still requires `schemas/train.schema.json` and `references/spec_template_train.yaml` to exist and parse. Use the packaged train schema for `automl_default_parameters`, `automl_disabled_parameters`, defaults, min/max bounds, enums, option weights, math conditions, dependencies, and popular parameters. Do not expect `~/tao-core` at runtime; maintainers regenerate schemas/templates before packaging the skill bank. |
| 28 | |
| 29 | ## Train Action Policy |
| 30 | |
| 31 | This model is AutoML-enabled at the model layer. Before handling any train-stage request, read `references/skill_info.yaml` and resolve the run override from either an explicit `automl_policy` value or the user's workflow request. Use `automl_policy: on` by default and only expose `on` / `off` in new launch prompts. Treat phrases like "turn off AutoML", "disable AutoML", "no HPO", or "plain training" as `automl_policy: off` for this run only. When `automl_policy: on`, `automl_enabled: true`, and both `schemas/train.schema.json` and `references/spec_template_train.yaml` are packaged, route the train action through `tao-skill-bank:tao-run-automl` by default with this model's `skill_dir`. Preserve workflow/application overrides for datasets, specs, output directories, GPU/platform settings, parent checkpoints, and `automl_policy`. Use direct model training only when `automl_policy: off` or the packaged train schema/template is missing; in the missing-schema case, report that AutoML is enabled but not runnable for this model until schemas are generated. |
| 32 | |
| 33 | Non-train actions such as `evaluate`, `inference`, `export`, and deploy flows stay in this model skill. The per-run `automl_policy` override does not change model metadata. |
| 34 | |
| 35 | ## Training Requirements |
| 36 | |
| 37 | The agent MUST read this section before generating any training or AutoML script for DINO. |
| 38 | |
| 39 | - **Dataset type:** object_detection |
| 40 | - **Formats:** coco, coco_raw |
| 41 | - **Accepted dataset intents:** training, evaluation, testing, calibration |
| 42 | - **Monitoring metric:** mAP50 for quick operational checks; `val_mAP` for |
| 43 | COCO/paper-style benchmark comparisons. |
| 44 | |
| 45 | **Required datasets — MUST resolve both:** |
| 46 | |
| 47 | | Dataset | Required | Why | |
| 48 | |---|---|---| |
| 49 | | Train dataset URI | Yes | Training data (COCO format) | |
| 50 | | Validation dataset URI | **Yes — ALWAYS** | DINO unconditionally builds a val dataloader. Omitting `val_data_sources` causes `FileNotFoundError` at startup regardless of the metric or workflow. If the user |
People who liked this also viewed
6 picksType
- hyperframes-cliUse the HyperFrames CLI development loop: init, add, catalog, capture, lint, check, snapshot, compare, grade-compare, preview, play, present, beats, keyframes, single or batch render, publish, cloud, SkillsJul 2026
- request-refactor-planSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026
- analyze-projectRigor Analyze / Rigor Audit read-only skill for deep learning research repositories. Use when the user wants to read and understand a repository, inspect model structure and training or inference entrSkillsMay 2026
- explore-codeRigor Improve implementation leaf skill for auditable candidate implementation in deep learning research repositories. Use when the researcher explicitly authorizes exploratory work on an isolated braSkillsMay 2026
- ubiquitous-languageSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026
- scaffold-exercisesSkills for Real Engineers. Straight from my .agents directory.SkillsJul 2026
