Tao Train Centerpose

CenterPose

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).

CenterPose for keypoint / pose estimation. Detects object centers and regresses keypoint locations. Used for 6-DoF object pose estimation.

Set model.backbone.pretrained_backbone_path.

For TAO Deploy TensorRT actions (gen_trt_engine, TensorRT evaluate, and TensorRT inference), use the deploy spec templates packaged in this skill's references/ folder with the spec_template_deploy_*.yaml prefix.

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 for an action requires schemas/<action>.schema.json and references/spec_template_<action>.yaml to exist and parse. Use the packaged selected-action 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

• Dataset type: centerpose
• Formats: default
• Training monitoring metrics: val_3DIoU, val_2DMPE
• Evaluate task metrics: test_3DIoU, test_2DMPE
• AutoML metric contract: use test_3DIoU with maximize direction for the required evaluation-backed baseline, every recommendation through eval_fn, best-model selection, and final evaluation. val_3DIoU is emitted by training but not by the evaluate action, so it is only suitable for an explicitly accepted training-proxy run without an impact baseline.

Per-Action Dataset Requirements

Typical Spec Overrides

Data source overrides are mandatory for every action — the agent MUST construct data source paths from the Per-Action Dataset Requirements table above and include them in spec_overrides.

TRAIN_DIR = "/path/to/extracted/train"
VAL_DIR = "/path/to/extracted/val"
TEST_DIR = "/path/to/extracted/test"
INFER_DIR = VAL_DIR
CAL_IMAGE_DIRS = ["/path/to/extracted/train/<sequence_or_image_dir>"]

train (mandatory data sources):

{
    "train.num_epochs": 30,
    "train.checkpoint_interval": 10,
    "train.validation_interval": 10,
    "train.num_gpus": 1,
    "dataset.category": "bike",
    "dataset.batch_size": 4,
    "dataset.train_data": TRAIN_DIR,
    "dataset.val_data": VAL_DIR,
}

evaluate (mandatory data sources):

{
    "dataset.category": "bike",
    "dataset.test_data": TEST_DIR,
}

inference (mandatory data sources):

{
    "dataset.category": "bike",
    "dataset.inference_data": INFER_DIR,
}

gen_trt_engine (mandatory data sources):

{
    "gen_trt_engine.tensorrt.calibration.cal_image_dir": CAL_IMAGE_DIRS,
}

Eval Dataset

Optional. Val and test datasets are provided as separate tarballs. Training writes val_3DIoU/val_2DMPE KPIs, while evaluate writes test_3DIoU/test_2DMPE; do not configure an evaluate-backed AutoML callback to extract the training-prefixed names.

Important Parameters

• dataset.num_classes: Number of object categories. Default 1.
• dataset.num_joints: Number of keypoints per object. Fixed at 8 (bbox keypoints). Valid range: exactly 8.
• dataset.input_res: Input resolution. Fixed at 512. Output resolution fixed at 128.
• dataset.category: Object category name. Default "cereal_box".
• model.backbone.model_type: Default fan_small. Backbone options limited in schema.
• train.optim.lr: Learning rate. Default 6e-5. MultiStep scheduler with lr_steps=[90, 120], lr_decay=0.1.
• train.loss_config: Rich loss config with toggles: mse_loss, obj_scale, obj_scale_uncertainty, hps_uncertainty, reg_bbox, hm_hp. Weights: wh_weight=0.1, off_weight=1, hp_weight=1.
• inference.use_pnp: Use PnP for 6-DoF pose. Default True. Requires camera intrinsics (focal_length_x/y, principle_point_x/y).
• export.input_width: Export input size. Fixed at 512x512. opset_version=16.

Multi-GPU / Multi-Node

Launch method: Lightning-managed (single python process, Lightning spawns workers).

• Strategy: auto (Lightning picks the best strategy automatically)
• No explicit num_nodes or distributed_strategy config — single-node only
• No sync_batchnorm

Export / TRT Defaults

• Export input: 512x512 (fixed), opset 16
• TRT data types: FP32, FP16, INT8
• TRT opt_batch_size: 4, max_batch_size: 8