[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7A7872MPSDFBJC2YQQADW60":3,"skill-same:01M7A7872MPSDFBJC2YQQADW60":121},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":31,"commitSha":32,"skillDirectory":33,"forksCount":34,"repositoryHtmlUrl":35,"repositoryHomepageUrl":36,"ownerHtmlUrl":37,"ownerBlogUrl":38,"ownerDescription":39,"updatedAt":40,"install":41,"resources":54},"01M7A7872MPSDFBJC2YQQADW60","tao-train-mask-grounding-dino",null,"Mask Grounding DINO for grounded instance segmentation. Extends Grounding DINO with a mask-prediction head for open-set segmentation guided by text prompts. Use when training, evaluating, exporting, quantizing, or running inference for a TAO Mask-Grounding-DINO model. Trigger phrases include \"train Mask Grounding DINO\", \"open-vocabulary segmentation\", \"text-prompted instance segmentation\", \"grounded mask DETR\".","themusebook\u002Fskills","themusebook","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F349927247?v=4","01M7A5C1RV5S3V5FF6ZACVDNRN","musebook","01M7A5C1SFC3KRZAXNN60XJ7KM",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: tao-train-mask-grounding-dino\ndescription: Mask Grounding DINO for grounded instance segmentation. Extends Grounding DINO with a mask-prediction head for\n  open-set segmentation guided by text prompts. Use when training, evaluating, exporting, quantizing, or running inference for\n  a TAO Mask-Grounding-DINO model. Trigger phrases include \"train Mask Grounding DINO\", \"open-vocabulary segmentation\",\n  \"text-prompted instance segmentation\", \"grounded mask DETR\".\nlicense: Apache-2.0\ncompatibility: Requires docker + nvidia-container-toolkit.\nmetadata:\n  version: \"0.1.0\"\n  author: NVIDIA Corporation\nallowed-tools: Read Bash\ntags:\n- segmentation\n---\n\n# Mask Grounding DINO\n\nMask Grounding DINO for grounded instance segmentation. Extends Grounding DINO with mask prediction head for open-set segmentation guided by text prompts.\n\nSet train.pretrained_model_path for full model weights.\n\nFor TAO Deploy TensorRT actions (`gen_trt_engine`, TensorRT `evaluate`, and TensorRT `inference`), read `references\u002Ftao-deploy-mask-grounding-dino.md` first. Deploy spec templates live in this skill's `references\u002F` folder with the `spec_template_deploy_*.yaml` prefix.\n\n## Dataclass Schemas\n\nGenerated TAO Core schemas are packaged in `schemas\u002F\u003Caction>.schema.json`, with `schemas\u002Fmanifest.json` listing available actions. Each generated schema also emits `references\u002Fspec_template_\u003Caction>.yaml` from the schema top-level `default` field. AutoML enablement is declared at the model layer in `references\u002Fskill_info.yaml` via `automl_enabled`. Runnable AutoML still requires `schemas\u002Ftrain.schema.json` and `references\u002Fspec_template_train.yaml` to exist and parse. Use the packaged train schema for `automl_default_parameters`, `automl_disabled_parameters`, defaults, min\u002Fmax bounds, enums, option weights, math conditions, dependencies, and popular parameters. Do not expect `~\u002Ftao-core` at runtime; maintainers regenerate schemas\u002Ftemplates before packaging the skill bank.\n\n## Train Action Policy\n\nThis model is AutoML-enabled at the model layer. Before handling any train-stage request, read `references\u002Fskill_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` \u002F `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\u002Ftrain.schema.json` and `references\u002Fspec_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\u002Fapplication overrides for datasets, specs, output directories, GPU\u002Fplatform settings, parent checkpoints, and `automl_policy`. Use direct model training only when `automl_policy: off` or the packaged train schema\u002Ftemplate is missing; in the missing-schema case, report that AutoML is enabled but not runnable for this model until schemas are generated.\n\nNon-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.\n\n## Training Requirements\n\n- **Dataset type:** segmentation\n- **Formats:** odvg, coco, coco_raw\n- **Monitoring metric:** val_loss\n\n### Per-Action Dataset Requirements\n\n| Action | Spec Key | Source | Files | List? |\n|---|---|---|---|---|\n| evaluate | dataset.test_data_sources | eval_dataset | image_dir: images.tar.gz, json_file: annotations.json | No |\n| evaluate | dataset.test_data_sources.data_type | eval_dataset | OD | No |\n| inference | dataset.infer_data_sources | inference_dataset | image_dir: images.tar.gz, captions: text prompts | No |\n| inference | dataset.infer_data_sources.data_type | inference_dataset | OD | No |\n| quantize | dataset.train_data_sources | train_datasets | image_dir: images.tar.gz, json_file: annotations_odvg.jsonl, label_map: annotations_odvg_labelmap.json | Yes |\n| quantize | dataset.val_data_sources | eval_dataset | image_dir: images.tar.gz, json_file: annotations.json | No |\n| quantize | dataset.val_data_sources.data_type | eval_dataset | OD | No |\n| quantize | dataset.quant_calibration_data_sources | train_datasets | image_dir: images.tar.gz, json_file: annotations_odvg.jsonl, label_map: annotations_odvg_labelmap.json | No |\n| train | dataset.train_data_sources | train_datasets | image_dir: images.tar.gz, json_file: annotations_odvg.jsonl, label_map: annotations_odvg_labelmap.json | Yes |\n| train | dataset.val_data_sources | eval_dataset | image_dir: images.tar.gz, json_file: annotations.json | No |\n| train | dataset.val_data_sources.data_type | eval_dataset | OD | No |\n\n### Typical Spec Overrides\n\nData 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`.\n\n```python\nS3_TRAIN = \"s3:\u002F\u002Fbucket\u002Fdata\u002Ftrain\"\nS3_EVAL = \"s3:\u002F\u002Fbucket\u002Fdata\u002Feval\"\n```\n\n**train (mandatory data sources):**\n```python\n{\n    \"train.num_gpus\": 1,\n    \"train.num_epochs\": 10,\n    \"train.checkpoint_interval\": 10,\n    \"train.validation_interval\": 10,\n    \"dataset.val_data_sources.data_type\": \"OD\",\n    \"model.num_region_queries\": 100,\n    \"dataset.train_data_sources\": [{\"image_dir\": f\"{S3_TRAIN}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_TRAIN}\u002Fannotations_odvg.jsonl\", \"label_map\": f\"{S3_TRAIN}\u002Fannotations_odvg_labelmap.json\"}],\n    \"dataset.val_data_sources\": {\"image_dir\": f\"{S3_EVAL}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_EVAL}\u002Fannotations.json\"},\n}\n```\n\n**evaluate (mandatory data sources):**\n```python\n{\n    \"evaluate.checkpoint\": \"\u003Cselected train\u002FAutoML checkpoint>\",\n    \"dataset.test_data_sources.data_type\": \"OD\",\n    \"dataset.test_data_sources\": {\"image_dir\": f\"{S3_EVAL}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_EVAL}\u002Fannotations.json\"},\n}\n```\n\n**inference (mandatory data sources):**\n```python\n{\n    \"inference.checkpoint\": \"\u003Cselected train\u002FAutoML checkpoint>\",\n    \"dataset.infer_data_sources.data_type\": \"OD\",\n    \"dataset.infer_data_sources\": {\"image_dir\": f\"{S3_EVAL}\u002Fimages.tar.gz\", \"captions\": [\"person\", \"bicycle\", \"car\"]},\n}\n```\n\n**quantize (mandatory data sources):**\n```python\n{\n    \"dataset.train_data_sources\": [{\"image_dir\": f\"{S3_TRAIN}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_TRAIN}\u002Fannotations_odvg.jsonl\", \"label_map\": f\"{S3_TRAIN}\u002Fannotations_odvg_labelmap.json\"}],\n    \"dataset.val_data_sources\": {\"image_dir\": f\"{S3_EVAL}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_EVAL}\u002Fannotations.json\"},\n    \"dataset.quant_calibration_data_sources\": {\"image_dir\": f\"{S3_TRAIN}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_TRAIN}\u002Fannotations_odvg.jsonl\", \"label_map\": f\"{S3_TRAIN}\u002Fannotations_odvg_labelmap.json\"},\n}\n```\n## Eval Dataset\n\nOptional. Validation uses COCO-format annotations even when training uses ODVG.\n\n## Important Parameters\n\n- **model.backbone**: Default swin_tiny_224_1k. Same backbone options as Grounding DINO.\n- **train.optim.lr**: Learning rate. Default 2e-4. lr_backbone 2e-5. Reuses GDINOTrainExpConfig — same training setup as Grounding DINO.\n- **model.num_queries**: Object queries. Default 900.\n- **model.enc_layers \u002F model.dec_layers**: Keep both at 6 for train\u002FAutoML\n  runs. The mask head asserts six decoder outputs during validation, so\n  copying Grounding DINO smoke overrides that reduce transformer layers causes\n  an immediate failure.\n- **AutoML metric note**: Use `metric=\"val_loss\"` with\n  `direction=\"minimize\"` for train-stage AutoML. The packaged train loop logs\n  validation loss scalars; it does not emit `[bbox] val_mAP@50` during the\n  train job.\n- **model.has_mask**: Enables mask prediction head. Default True. Adds mask\u002Fdice\u002Frela loss coefficients.\n- **model.num_region_queries**: Number of region queries for mask prediction. Default 100.\n- **model.loss_types**: Loss components. Default [labels, boxes, masks]. Includes mask_loss_coef, dice_loss_coef, rela_loss_coef.\n- **evaluate.ioi_threshold**: IoI threshold for mask evaluation. Default 0.5.\n- **evaluate.nms_threshold**: NMS threshold. Default 0.2.\n- **evaluate.text_threshold**: Text matching threshold. Default 0.3.\n- **dataset.has_mask**: Dataset includes mask annotations. Default True. val_data_sources default data_type is \"VG\".\n\n## Multi-GPU \u002F Multi-Node\n\n**Launch method:** Lightning-managed. Same DDP\u002FFSDP behavior as Grounding DINO.\n\n| Spec Key | Description | Default |\n|----------|-------------|---------|\n| `train.num_gpus` | Number of GPUs | 1 |\n| `train.gpu_ids` | GPU device indices | [0] |\n| `train.num_nodes` | Number of nodes | 1 |\n| `train.distributed_strategy` | `ddp` or `fsdp` | `ddp` |\n\n## Hardware\n\nMinimum 1 GPU(s), recommended 4 GPU(s). 24GB+ (A100 recommended) VRAM per GPU. Heavier than Grounding DINO due to mask prediction head. 24GB+ GPU memory recommended.\n\n## Error Patterns\n\n**CUDA out of memory**: Reduce batch_size. Mask prediction adds overhead on top of Grounding DINO.\n\n**Deploy schema error for `test_threshold`**: TAO Deploy uses\n`evaluate.text_threshold` and `inference.text_threshold`. Do not use\n`test_threshold` in deploy specs.\n\n**Deploy model shape mismatch**: Carry transformer and mask structure fields\nfrom export into deploy evaluate\u002Finference specs, including `model.num_queries`,\n`model.num_select`, `model.max_text_len`, `model.num_region_queries`, and\n`model.has_mask`. These values must match the ONNX model used to build the\nTensorRT engine.\n\n## Spec Param \u002F Parent Model Inference\n\nModel-specific inference mappings belong in this MD file, not in `config.json`. Generated runners should read this section and apply the mappings with SDK helpers before `create_job()`. This mirrors the old microservices `infer_params.py` flow.\n\nInference mappings from TAO Core `mask_grounding_dino.config.json`:\n\n| Action | Spec Field | Inference Function | Meaning |\n|---|---|---|---|\n| evaluate | `encryption_key` | `key` | encryption key |\n| evaluate | `evaluate.checkpoint` | `parent_model` | model file inferred from the parent job results folder |\n| evaluate | `evaluate.trt_engine` | `parent_model` | model file inferred from the parent job results folder |\n| evaluate | `results_dir` | `output_dir` | current job results directory |\n| export | `encryption_key` | `key` | encryption key |\n| export | `export.checkpoint` | `parent_model` | model file inferred from the parent job results folder |\n| export | `export.onnx_file` | `create_onnx_file` | output ONNX path |\n| export | `results_dir` | `output_dir` | current job results directory |\n| gen_trt_engine | `encryption_key` | `key` | encryption key |\n| gen_trt_engine | `gen_trt_engine.onnx_file` | `parent_model` | model file inferred from the parent job results folder |\n| gen_trt_engine | `gen_trt_engine.trt_engine` | `create_engine_file` | output TensorRT engine path |\n| gen_trt_engine | `results_dir` | `output_dir` | current job results directory |\n| inference | `encryption_key` | `key` | encryption key |\n| inference | `inference.checkpoint` | `parent_model` | model file inferred from the parent job results folder |\n| inference | `inference.trt_engine` | `parent_model` | model file inferred from the parent job results folder |\n| inference | `results_dir` | `output_dir` | current job results directory |\n| quantize | `encryption_key` | `key` | encryption key |\n| quantize | `quantize.model_path` | `parent_model` | model file inferred from the parent job results folder |\n| quantize | `results_dir` | `output_dir` | current job results directory |\n| train | `encryption_key` | `key` | encryption key |\n| train | `model.pretrained_backbone_path` | `ptm_if_no_resume_model` | PTM when no resume checkpoint exists |\n| train | `results_dir` | `output_dir` | current job results directory |\n| train | `train.pretrained_model_path` | `ptm_if_no_resume_model` | PTM when no resume checkpoint exists |\n| train | `train.resume_training_checkpoint_path` | `resume_model` | model file inferred from the current job results folder |\n\nFor `parent_model` or `parent_model_folder`, pass the upstream train\u002Fexport\u002FAutoML child job id as `parent_job_id`. The SDK lists the parent result folder, filters checkpoint artifacts, and returns the selected model file or folder. Do not add these mappings back to `config.json` and do not patch generated runner scripts to guess checkpoint paths.\n\nWhen selecting a Mask Grounding DINO checkpoint outside the SDK resolver, match\nthe intended epoch\u002Fstep artifact exactly, for example\n`model_epoch_000_step_00049.pth`. The `mask_gdino_model_latest.pth` symlink is\nvalid only when latest is explicitly requested. The parent PyTorch\n`mask_grounding_dino` CLI supports `train`, `evaluate`, `inference`, `export`,\nand `quantize`; run TensorRT engine generation, TensorRT inference, and\nTensorRT evaluation through `references\u002Ftao-deploy-mask-grounding-dino.md`.\n\n## Deployment\n\n- [tao-deploy-mask-grounding-dino](references\u002Ftao-deploy-mask-grounding-dino.md)\n","# Mask Grounding DINO\n\nMask Grounding DINO for grounded instance segmentation. Extends Grounding DINO with mask prediction head for open-set segmentation guided by text prompts.\n\nSet train.pretrained_model_path for full model weights.\n\nFor TAO Deploy TensorRT actions (`gen_trt_engine`, TensorRT `evaluate`, and TensorRT `inference`), read `references\u002Ftao-deploy-mask-grounding-dino.md` first. Deploy spec templates live in this skill's `references\u002F` folder with the `spec_template_deploy_*.yaml` prefix.\n\n## Dataclass Schemas\n\nGenerated TAO Core schemas are packaged in `schemas\u002F\u003Caction>.schema.json`, with `schemas\u002Fmanifest.json` listing available actions. Each generated schema also emits `references\u002Fspec_template_\u003Caction>.yaml` from the schema top-level `default` field. AutoML enablement is declared at the model layer in `references\u002Fskill_info.yaml` via `automl_enabled`. Runnable AutoML still requires `schemas\u002Ftrain.schema.json` and `references\u002Fspec_template_train.yaml` to exist and parse. Use the packaged train schema for `automl_default_parameters`, `automl_disabled_parameters`, defaults, min\u002Fmax bounds, enums, option weights, math conditions, dependencies, and popular parameters. Do not expect `~\u002Ftao-core` at runtime; maintainers regenerate schemas\u002Ftemplates before packaging the skill bank.\n\n## Train Action Policy\n\nThis model is AutoML-enabled at the model layer. Before handling any train-stage request, read `references\u002Fskill_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` \u002F `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\u002Ftrain.schema.json` and `references\u002Fspec_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\u002Fapplication overrides for datasets, specs, output directories, GPU\u002Fplatform settings, parent checkpoints, and `automl_policy`. Use direct model training only when `automl_policy: off` or the packaged train schema\u002Ftemplate is missing; in the missing-schema case, report that AutoML is enabled but not runnable for this model until schemas are generated.\n\nNon-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.\n\n## Training Requirements\n\n- **Dataset type:** segmentation\n- **Formats:** odvg, coco, coco_raw\n- **Monitoring metric:** val_loss\n\n### Per-Action Dataset Requirements\n\n| Action | Spec Key | Source | Files | List? |\n|---|---|---|---|---|\n| evaluate | dataset.test_data_sources | eval_dataset | image_dir: images.tar.gz, json_file: annotations.json | No |\n| evaluate | dataset.test_data_sources.data_type | eval_dataset | OD | No |\n| inference | dataset.infer_data_sources | inference_dataset | image_dir: images.tar.gz, captions: text prompts | No |\n| inference | dataset.infer_data_sources.data_type | inference_dataset | OD | No |\n| quantize | dataset.train_data_sources | train_datasets | image_dir: images.tar.gz, json_file: annotations_odvg.jsonl, label_map: annotations_odvg_labelmap.json | Yes |\n| quantize | dataset.val_data_sources | eval_dataset | image_dir: images.tar.gz, json_file: annotations.json | No |\n| quantize | dataset.val_data_sources.data_type | eval_dataset | OD | No |\n| quantize | dataset.quant_calibration_data_sources | train_datasets | image_dir: images.tar.gz, json_file: annotations_odvg.jsonl, label_map: annotations_odvg_labelmap.json | No |\n| train | dataset.train_data_sources | train_datasets | image_dir: images.tar.gz, json_file: annotations_odvg.jsonl, label_map: annotations_odvg_labelmap.json | Yes |\n| train | dataset.val_data_sources | eval_dataset | image_dir: images.tar.gz, json_file: annotations.json | No |\n| train | dataset.val_data_sources.data_type | eval_dataset | OD | No |\n\n### Typical Spec Overrides\n\nData 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`.\n\n```python\nS3_TRAIN = \"s3:\u002F\u002Fbucket\u002Fdata\u002Ftrain\"\nS3_EVAL = \"s3:\u002F\u002Fbucket\u002Fdata\u002Feval\"\n```\n\n**train (mandatory data sources):**\n```python\n{\n    \"train.num_gpus\": 1,\n    \"train.num_epochs\": 10,\n    \"train.checkpoint_interval\": 10,\n    \"train.validation_interval\": 10,\n    \"dataset.val_data_sources.data_type\": \"OD\",\n    \"model.num_region_queries\": 100,\n    \"dataset.train_data_sources\": [{\"image_dir\": f\"{S3_TRAIN}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_TRAIN}\u002Fannotations_odvg.jsonl\", \"label_map\": f\"{S3_TRAIN}\u002Fannotations_odvg_labelmap.json\"}],\n    \"dataset.val_data_sources\": {\"image_dir\": f\"{S3_EVAL}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_EVAL}\u002Fannotations.json\"},\n}\n```\n\n**evaluate (mandatory data sources):**\n```python\n{\n    \"evaluate.checkpoint\": \"\u003Cselected train\u002FAutoML checkpoint>\",\n    \"dataset.test_data_sources.data_type\": \"OD\",\n    \"dataset.test_data_sources\": {\"image_dir\": f\"{S3_EVAL}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_EVAL}\u002Fannotations.json\"},\n}\n```\n\n**inference (mandatory data sources):**\n```python\n{\n    \"inference.checkpoint\": \"\u003Cselected train\u002FAutoML checkpoint>\",\n    \"dataset.infer_data_sources.data_type\": \"OD\",\n    \"dataset.infer_data_sources\": {\"image_dir\": f\"{S3_EVAL}\u002Fimages.tar.gz\", \"captions\": [\"person\", \"bicycle\", \"car\"]},\n}\n```\n\n**quantize (mandatory data sources):**\n```python\n{\n    \"dataset.train_data_sources\": [{\"image_dir\": f\"{S3_TRAIN}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_TRAIN}\u002Fannotations_odvg.jsonl\", \"label_map\": f\"{S3_TRAIN}\u002Fannotations_odvg_labelmap.json\"}],\n    \"dataset.val_data_sources\": {\"image_dir\": f\"{S3_EVAL}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_EVAL}\u002Fannotations.json\"},\n    \"dataset.quant_calibration_data_sources\": {\"image_dir\": f\"{S3_TRAIN}\u002Fimages.tar.gz\", \"json_file\": f\"{S3_TRAIN}\u002Fannotations_odvg.jsonl\", \"label_map\": f\"{S3_TRAIN}\u002Fannotations_odvg_labelmap.json\"},\n}\n```\n## Eval Dataset\n\nOptional. Validation uses COCO-format annotations even when training uses ODVG.\n\n## Important Parameters\n\n- **model.backbone**: Default swin_tiny_224_1k. Same backbone options as Grounding DINO.\n- **train.optim.lr**: Learning rate. Default 2e-4. lr_backbone 2e-5. Reuses GDINOTrainExpConfig — same training setup as Grounding DINO.\n- **model.num_queries**: Object queries. Default 900.\n- **model.enc_layers \u002F model.dec_layers**: Keep both at 6 for train\u002FAutoML\n  runs. The mask head asserts six decoder outputs during validation, so\n  copying Grounding DINO smoke overrides that reduce transformer layers causes\n  an immediate failure.\n- **AutoML metric note**: Use `metric=\"val_loss\"` with\n  `direction=\"minimize\"` for train-stage AutoML. The packaged train loop logs\n  validation loss scalars; it does not emit `[bbox] val_mAP@50` during the\n  train job.\n- **model.has_mask**: Enables mask prediction head. Default True. Adds mask\u002Fdice\u002Frela loss coefficients.\n- **model.num_region_queries**: Number of region queries for mask prediction. Default 100.\n- **model.loss_types**: Loss components. Default [labels, boxes, masks]. Includes mask_loss_coef, dice_loss_coef, rela_loss_coef.\n- **evaluate.ioi_threshold**: IoI threshold for mask evaluation. Default 0.5.\n- **evaluate.nms_threshold**: NMS threshold. Default 0.2.\n- **evaluate.text_threshold**: Text matching threshold. Default 0.3.\n- **dataset.has_mask**: Dataset includes mask annotations. Default True. val_data_sources default data_type is \"VG\".\n\n## Multi-GPU \u002F Multi-Node\n\n**Launch method:** Lightning-managed. Same DDP\u002FFSDP behavior as Grounding DINO.\n\n| Spec Key | Description | Default |\n|----------|-------------|---------|\n| `train.num_gpus` | Number of GPUs | 1 |\n| `train.gpu_ids` | GPU device indices | [0] |\n| `train.num_nodes` | Number of nodes | 1 |\n| `train.distributed_strategy` | `ddp` or `fsdp` | `ddp` |\n\n## Hardware\n\nMinimum 1 GPU(s), recommended 4 GPU(s). 24GB+ (A100 recommended) VRAM per GPU. Heavier than Grounding DINO due to mask prediction head. 24GB+ GPU memory recommended.\n\n## Error Patterns\n\n**CUDA out of memory**: Reduce batch_size. Mask prediction adds overhead on top of Grounding DINO.\n\n**Deploy schema error for `test_threshold`**: TAO Deploy uses\n`evaluate.text_threshold` and `inference.text_threshold`. Do not use\n`test_threshold` in deploy specs.\n\n**Deploy model shape mismatch**: Carry transformer and mask structure fields\nfrom export into deploy evaluate\u002Finference specs, including `model.num_queries`,\n`model.num_select`, `model.max_text_len`, `model.num_region_queries`, and\n`model.has_mask`. These values must match the ONNX model used to build the\nTensorRT engine.\n\n## Spec Param \u002F Parent Model Inference\n\nModel-specific inference mappings belong in this MD file, not in `config.json`. Generated runners should read this section and apply the mappings with SDK helpers before `create_job()`. This mirrors the old microservices `infer_params.py` flow.\n\nInference mappings from TAO Core `mask_grounding_dino.config.json`:\n\n| Action | Spec Field | Inference Function | Meaning |\n|---|---|---|---|\n| evaluate | `encryption_key` | `key` | encryption key |\n| evaluate | `evaluate.checkpoint` | `parent_model` | model file inferred from the parent job results folder |\n| evaluate | `evaluate.trt_engine` | `parent_model` | model file inferred from the parent job results folder |\n| evaluate | `results_dir` | `output_dir` | current job results directory |\n| export | `encryption_key` | `key` | encryption key |\n| export | `export.checkpoint` | `parent_model` | model file inferred from the parent job results folder |\n| export | `export.onnx_file` | `create_onnx_file` | output ONNX path |\n| export | `results_dir` | `output_dir` | current job results directory |\n| gen_trt_engine | `encryption_key` | `key` | encryption key |\n| gen_trt_engine | `gen_trt_engine.onnx_file` | `parent_model` | model file inferred from the parent job results folder |\n| gen_trt_engine | `gen_trt_engine.trt_engine` | `create_engine_file` | output TensorRT engine path |\n| gen_trt_engine | `results_dir` | `output_dir` | current job results directory |\n| inference | `encryption_key` | `key` | encryption key |\n| inference | `inference.checkpoint` | `parent_model` | model file inferred from the parent job results folder |\n| inference | `inference.trt_engine` | `parent_model` | model file inferred from the parent job results folder |\n| inference | `results_dir` | `output_dir` | current job results directory |\n| quantize | `encryption_key` | `key` | encryption key |\n| quantize | `quantize.model_path` | `parent_model` | model file inferred from the parent job results folder |\n| quantize | `results_dir` | `output_dir` | current job results directory |\n| train | `encryption_key` | `key` | encryption key |\n| train | `model.pretrained_backbone_path` | `ptm_if_no_resume_model` | PTM when no resume checkpoint exists |\n| train | `results_dir` | `output_dir` | current job results directory |\n| train | `train.pretrained_model_path` | `ptm_if_no_resume_model` | PTM when no resume checkpoint exists |\n| train | `train.resume_training_checkpoint_path` | `resume_model` | model file inferred from the current job results folder |\n\nFor `parent_model` or `parent_model_folder`, pass the upstream train\u002Fexport\u002FAutoML child job id as `parent_job_id`. The SDK lists the parent result folder, filters checkpoint artifacts, and returns the selected model file or folder. Do not add these mappings back to `config.json` and do not patch generated runner scripts to guess checkpoint paths.\n\nWhen selecting a Mask Grounding DINO checkpoint outside the SDK resolver, match\nthe intended epoch\u002Fstep artifact exactly, for example\n`model_epoch_000_step_00049.pth`. The `mask_gdino_model_latest.pth` symlink is\nvalid only when latest is explicitly requested. The parent PyTorch\n`mask_grounding_dino` CLI supports `train`, `evaluate`, `inference`, `export`,\nand `quantize`; run TensorRT engine generation, TensorRT inference, and\nTensorRT evaluation through `references\u002Ftao-deploy-mask-grounding-dino.md`.\n\n## Deployment\n\n- [tao-deploy-mask-grounding-dino](references\u002Ftao-deploy-mask-grounding-dino.md)\n",{"name":5,"tags":24,"license":15,"metadata":26,"description":7,"allowed-tools":29,"compatibility":30},[25],"segmentation",{"author":27,"version":28},"NVIDIA Corporation","0.1.0","Read Bash","Requires docker + nvidia-container-toolkit.","main","1c523aba5cdf867a79f078847075492d7778092a","public\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino",1,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills","https:\u002F\u002Fskills.x402.wtf","https:\u002F\u002Fgithub.com\u002Fthemusebook","https:\u002F\u002Fmusebook.trade","The official Github of musebook","2026-11-02 23:42:26",{"source":8,"skillName":5,"commands":42,"prompt":52,"zipAvailable":53},[43,46,49],{"tool":44,"command":45},"npx","npx skills add themusebook\u002Fskills --skill tao-train-mask-grounding-dino",{"tool":47,"command":48},"bunx","bunx skills add themusebook\u002Fskills --skill tao-train-mask-grounding-dino",{"tool":50,"command":51},"pnpm","pnpm dlx skills add themusebook\u002Fskills --skill tao-train-mask-grounding-dino","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002FSKILL.md",true,[55,61,65,69,73,77,81,85,89,93,97,101,105,109,113,117],{"path":56,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":59,"sourceHtmlUrl":60},"BENCHMARK.md","REFERENCE","参考文档",4136,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002FBENCHMARK.md",{"path":62,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":63,"sourceHtmlUrl":64},"metadata.json",659,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Fmetadata.json",{"path":66,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":67,"sourceHtmlUrl":68},"references\u002Fskill_info.yaml",6309,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fskill_info.yaml",{"path":70,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":71,"sourceHtmlUrl":72},"references\u002Fspec_template_deploy_evaluate.yaml",567,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fspec_template_deploy_evaluate.yaml",{"path":74,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":75,"sourceHtmlUrl":76},"references\u002Fspec_template_deploy_gen_trt_engine.yaml",365,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fspec_template_deploy_gen_trt_engine.yaml",{"path":78,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":79,"sourceHtmlUrl":80},"references\u002Fspec_template_deploy_inference.yaml",1651,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fspec_template_deploy_inference.yaml",{"path":82,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":83,"sourceHtmlUrl":84},"references\u002Fspec_template_evaluate.yaml",3612,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fspec_template_evaluate.yaml",{"path":86,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":87,"sourceHtmlUrl":88},"references\u002Fspec_template_export.yaml",3645,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fspec_template_export.yaml",{"path":90,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":91,"sourceHtmlUrl":92},"references\u002Fspec_template_gen_trt_engine.yaml",3682,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fspec_template_gen_trt_engine.yaml",{"path":94,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":95,"sourceHtmlUrl":96},"references\u002Fspec_template_inference.yaml",3692,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fspec_template_inference.yaml",{"path":98,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":99,"sourceHtmlUrl":100},"references\u002Fspec_template_quantize.yaml",3418,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fspec_template_quantize.yaml",{"path":102,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":103,"sourceHtmlUrl":104},"references\u002Fspec_template_train.yaml",3400,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Fspec_template_train.yaml",{"path":106,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":107,"sourceHtmlUrl":108},"references\u002Ftao-deploy-mask-grounding-dino.md",5683,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Ftao-deploy-mask-grounding-dino.md",{"path":110,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":111,"sourceHtmlUrl":112},"references\u002Ftao-deploy-mask-grounding-dino.skill_info.yaml",3395,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Freferences\u002Ftao-deploy-mask-grounding-dino.skill_info.yaml",{"path":114,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":115,"sourceHtmlUrl":116},"skill-card.md",3794,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Fskill-card.md",{"path":118,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":119,"sourceHtmlUrl":120},"verification.json",3870,"https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fpublic\u002Fapi\u002Fskills\u002Fnvidia\u002Ftao-train-mask-grounding-dino\u002Fverification.json",[122,131,138,145,152,159],{"skillNo":123,"name":124,"nameZh":6,"description":125,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":126,"licenseFileUrl":127,"sourceHtmlUrl":128,"categories":129,"keywords":130,"publishedAt":19},"01M7AB34SPKVSDRRG009PM1CZH","youtube-clipper","YouTube 视频智能剪辑工具。下载视频和字幕，AI 分析生成精细章节（几分钟级别）， 用户选择片段后自动剪辑、翻译字幕为中英双语、烧录字幕到视频，并生成总结文案。 使用场景：当用户需要剪辑 YouTube 视频、生成短视频片段、制作双语字幕版本时。 关键词：视频剪辑、YouTube、字幕翻译、双语字幕、视频下载、clip video","MIT","https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube-clipper\u002FLICENSE","https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fyoutube-clipper\u002FSKILL.md",[],[],{"skillNo":132,"name":133,"nameZh":6,"description":134,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":126,"licenseFileUrl":6,"sourceHtmlUrl":135,"categories":136,"keywords":137,"publishedAt":19},"01M7AB18BH6N7KR7939HRNWVMQ","viem-integration","Integrate EVM blockchains using viem. Use when user says \"read blockchain data\", \"send transaction\", \"interact with smart contract\", \"connect to Ethereum\", \"use viem\", \"use wagmi\", \"wallet integration\", \"viem setup\", or mentions blockchain\u002FEVM development with TypeScript.","https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fviem-integration\u002FSKILL.md",[],[],{"skillNo":139,"name":140,"nameZh":6,"description":141,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":126,"licenseFileUrl":6,"sourceHtmlUrl":142,"categories":143,"keywords":144,"publishedAt":19},"01M7AB12GHGP48WCNXDQQ9222H","v4-security-foundations","Security-first Uniswap v4 hook development. Use when user mentions \"v4 hooks\", \"hook security\", \"PoolManager\", \"beforeSwap\", \"afterSwap\", or asks about V4 hook best practices, vulnerabilities, or audit requirements.","https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fv4-security-foundations\u002FSKILL.md",[],[],{"skillNo":146,"name":147,"nameZh":6,"description":148,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":126,"licenseFileUrl":6,"sourceHtmlUrl":149,"categories":150,"keywords":151,"publishedAt":19},"01M7AB0XM3PSR7V3Q2ZRR0Q0FB","v4-sdk-integration","App-layer SDK guide for building swap and liquidity experiences directly with the Uniswap v4 SDK. Use when user asks about \"v4 sdk\", \"uniswap v4\", \"v4 swap\", \"v4 liquidity\", \"PoolManager\", \"V4Planner\", \"StateView\", \"PositionManager\", \"pool state\", \"v4 position\", \"uniswap sdk\", or when building swap\u002Fliquidity UX directly with SDKs rather than via the Trading API.","https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fv4-sdk-integration\u002FSKILL.md",[],[],{"skillNo":153,"name":154,"nameZh":6,"description":155,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":126,"licenseFileUrl":6,"sourceHtmlUrl":156,"categories":157,"keywords":158,"publishedAt":19},"01M7AB0SGA93ASA588DJW4P9WT","v4-hook-generator","Generate Uniswap v4 hook contracts via OpenZeppelin MCP. Use when building custom swap logic, async swaps, hook-owned liquidity, custom curves, dynamic fees, MEV protection, limit orders, or oracle hooks.","https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Fv4-hook-generator\u002FSKILL.md",[],[],{"skillNo":160,"name":161,"nameZh":6,"description":162,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":126,"licenseFileUrl":6,"sourceHtmlUrl":163,"categories":164,"keywords":165,"publishedAt":19},"01M7AAZYH75H4CS03WSW2EZ8DM","testing","For testing with Light Protocol programs and clients on localnet, devnet, and mainnet validation.","https:\u002F\u002Fgithub.com\u002Fthemusebook\u002Fskills\u002Fblob\u002Fmain\u002Fskills\u002Ftesting\u002FSKILL.md",[],[]]