Tao Train Dino

DINO

Standalone install? If this session was not initialized by the TAO skill bank plugin, run the tao-setup skill first (host preflight, credentials, cross-skill discovery).

DINO (DETR with Improved DeNoising Anchor Boxes) for 2D object detection. Transformer-based detector with denoising training, multi-scale features, and optional distillation support.

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.

When To Use

Train, evaluate, export, distill, quantize, or run inference for a TAO DINO 2D object detector.

For TAO Deploy TensorRT actions (gen_trt_engine, TensorRT evaluate, and TensorRT inference), read references/tao-deploy-dino.md first. Deploy spec templates live in this skill's references/ folder with the spec_template_deploy_*.yaml prefix.

Reference Map

• 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.
• references/dino-actions-errors.md — important parameters, default values, evaluate/export defaults, hardware, and the full error-pattern catalog.
• references/dino-tuning-multigpu.md — full AutoML/HPO notes (metrics, hyperparameters, extractor) and multi-GPU spec consistency.
• references/tao-deploy-dino.md — TensorRT deploy workflow.
• references/detailed-guide.md — map to the detailed model guide.

Dataclass Schemas

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.

Train Action Policy

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.

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.

Training Requirements

The agent MUST read this section before generating any training or AutoML script for DINO.

• Dataset type: object_detection
• Formats: coco, coco_raw
• Accepted dataset intents: training, evaluation, testing, calibration
• AutoML metric contract: for the default evaluation-backed workflow, use test_mAP50 with maximize direction. Use test_mAP only when the user explicitly requests COCO/paper-style mAP.
• Training monitoring metrics: val_mAP50 (logged as Validation mAP50) for quick operational checks; val_mAP for COCO/paper-style benchmark comparisons.
• Evaluate action metrics: test_mAP50 for AP50 and test_mAP for COCO mAP. AutoML workflows that score recommendations with the standalone evaluate action must use the corresponding test_* KPI.

Required datasets — MUST resolve both:

Required inputs before generating any training spec:

1. Train dataset URI — S3 path to COCO-format training data
2. Validation dataset URI — S3 path to COCO-format val data (can be same as train)
3. `num_classes` — How many object classes? Default 91 (COCO). Must be >= max(category_id) + 1. Too low causes CUDA error: device-side assert triggered.

Resolve these from the user request or the default profile below. Prompt only for values that are still missing after applying the profile rules.

Bankable local default profile for DINO AutoML smoke runs:

Use this profile only when the user asks to run DINO AutoML and does not provide dataset or class-count inputs. This profile is intentionally small and local to this skill bank; it is for smoke/iteration runs, not a production benchmark. Do not search previous runners, logs, session state, shell history, or the home directory to recover these values.

DINO_AUTOML_PROFILE = {
    "train_dataset_uri": "s3://nvcf-storage-handling/data/tao_od_synthetic_subset_train_no_convert",
    "validation_dataset_uri": "s3://nvcf-storage-handling/data/tao_od_synthetic_subset_val_no_convert",
    "object_classes": 4,
    "dataset_num_classes": 5,
    "image_archive": "images.tar.gz",
    "annotation_file": "annotations.json",
    "max_recommendations": 10,
    "train_num_epochs": 10,
    "train_checkpoint_interval": 10,
    "train_validation_interval": 1,
    "train_num_gpus": 1,
}

If the user supplies any dataset URI or class-count value, prefer the user value and ask for any remaining required DINO value. Do not partially mix a user's custom dataset with this profile's class count unless the user confirms it.

Do not prompt for image layout for the standard DINO dataset. The standard TAO DINO dataset artifact is images.tar.gz plus annotations.json. Use images.tar.gz in the remote image_dir spec override. The SDK downloads the archive and rewrites the runtime spec to the extracted folder named after the archive stem (images.tar.gz -> images). Only deviate if the user explicitly provides a different image artifact name.

Core Workflow

DINO supports train, evaluate, export, distill, quantize, and inference. Data-source overrides are mandatory for every action — DINO's config.json has empty data_sources because the runner cannot auto-resolve array-of-objects spec keys. The agent MUST construct data source paths and include them in spec_overrides.

See references/dino-data-specs.md for the per-action dataset requirements table, the standard dataset artifact (images.tar.gz + annotations.json) and runtime folder rewrite rules, and the complete per-action spec_overrides examples for train, evaluate, export, deploy/gen_trt_engine, inference, quantize, and distill — including checkpoint inference via parent_model, the results_dir/train/ checkpoint location, and the distillation FAN-teacher / student rules.

Important Parameters And Defaults

Default values: num_epochs=10, batch_size=4, learning_rate=2e-4,