[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7CQQ1R0AMADC8V8H04W4TH1":3,"skill-same:01M7CQQ1R0AMADC8V8H04W4TH1":61},{"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":33,"commitSha":34,"skillDirectory":35,"forksCount":36,"repositoryHtmlUrl":37,"repositoryHomepageUrl":6,"ownerHtmlUrl":38,"ownerBlogUrl":6,"ownerDescription":39,"updatedAt":40,"install":41,"resources":54},"01M7CQQ1R0AMADC8V8H04W4TH1","tao-train-foundation-stereo",null,"Stereo depth estimation using FoundationStereo. Predicts disparity maps from stereo image pairs for 3D reconstruction. Use when training, evaluating, exporting, or running inference for a TAO FoundationStereo model. Trigger phrases include \"train stereo depth\", \"FoundationStereo\", \"stereo disparity estimation\", \"3D reconstruction from stereo\".","om-scogo\u002Fskillsh-scraper","om-scogo","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F160466840?v=4","01M79AWB6DFVCQ7RK70CS34B96","Om Achrekar","01M79AWB6ZM9SKR8KDH1WSSM22",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fnvidia-skills\u002Ftao-train-foundation-stereo\u002FSkill.md",[],[],"2026-11-15 14:19:17",false,"---\nname: tao-train-foundation-stereo\ndescription: Stereo depth estimation using FoundationStereo. Predicts disparity maps from stereo image pairs for 3D\n  reconstruction. Use when training, evaluating, exporting, or running inference for a TAO FoundationStereo model. Trigger\n  phrases include \"train stereo depth\", \"FoundationStereo\", \"stereo disparity estimation\", \"3D reconstruction from stereo\".\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- stereo\n- depth\n- estimation\n---\n\n# Depth Net Stereo\n\nStereo depth estimation using FoundationStereo architecture. Predicts disparity maps from stereo image pairs for 3D reconstruction.\n\nUses pretrained Depth Anything v2 and EdgeNeXt encoders. Set `model.stereo_backbone.depth_anything_v2_pretrained_path` and `model.stereo_backbone.edgenext_pretrained_path`.\n\nThe mono and stereo skills both invoke the unified TAO `depth_net` CLI inside the container; the mono\u002Fstereo family is selected via `model.model_type` (e.g., `FoundationStereo`).\n\nPyT actions packaged by this model skill: `train`, `evaluate`, `inference`, `export`, and `quantize`. The PyT `depth_net` entrypoint does not accept a `gen_trt_engine` action in the current TAO image; build TensorRT engines only through the deploy workflow.\n\nFor TAO Deploy TensorRT actions (`gen_trt_engine`, TensorRT `evaluate`, and TensorRT `inference`), read `references\u002Ftao-deploy-foundation-stereo.md` first. The deploy spec template lives in this skill's `references\u002Fspec_template_deploy.yaml`.\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## Workflow\n\n### Prerequisites — data accessibility\n\nYour dataset (left + right images + GT disparity) must be reachable from inside the container:\n- **SDK runner**: place files at the S3 paths the runner resolves (the `S3_TRAIN` \u002F `S3_EVAL` placeholders shown in the spec overrides). The runner handles S3 → container-path mounting transparently.\n- **Direct `docker run`** (e.g. local testing): mount the host dataset root read-only at the same in-container path:\n\n```\ndocker run ... -v \u003Chost_data_root>:\u003Chost_data_root>:ro \u003Ccontainer> ...\n```\n\nThe same accessibility requirement applies to the `\u003Coutput_dir>` written by all actions.\n\n### Step 1 — Annotation file\n\nPer-line annotation file referenced by `data_sources[*].data_file`:\n\n| Columns | Format | Use |\n|---|---|---|\n| 2 | `\u003Cleft> \u003Cright>` | Stereo inference (no GT) |\n| 3 | `\u003Cleft> \u003Cright> \u003Cdisparity>` | Stereo with GT |\n| 4 | `\u003Cleft> \u003Cright> \u003Cdisparity> \u003Cocclusion_mask>` | Stereo with GT and occlusion mask |\n\nIf you already have one, point to it. Otherwise generate via `depth_net convert`:\n\n```\ndepth_net convert -e \u003Cconvert_spec.yaml>\n```\n\n`convert_spec.yaml` template (stereo):\n\n```yaml\nresults_dir: \u003Cdirectory where generated annotation files are written>\ndata_root: \u003Cdirectory whose immediate children are scene folders that contain your image+depth files; convert walks data_root recursively but expects per-scene subdirectories at one level below>\nimage_dir_pattern: [\u003Csubstring matching left image paths>]\nright_dir_pattern: [\u003Csubstring matching right image paths>]\ndepth_dir_pattern: [\u003Csubstring matching GT disparity paths>]\nnocc_dir_pattern: []                 # optional, occlusion mask paths\nimage_extension: '.png'  # always include the leading dot\ndepth_extension: '.png'  # form must match image_extension (the swap is a substring replace)\nnocc_extension: ''\nsplit_ratio: 0.0        # 0.0\u002F1.0 = test-only; 0.8 = 80\u002F20 train+val\n```\n\n`convert` walks `data_root` recursively, selects paths whose path-string contains *all* substrings in `image_dir_pattern` (AND-filter), then derives right \u002F depth \u002F mask paths by replacing `image_dir_pattern[0]` with the corresponding pattern's first element plus extension swap. Inspect your dataset's directory layout and identify the substrings distinguishing left, right, and GT (e.g. `im0` vs `im1` vs `disp0GT` for Middlebury).\n\n### Step 2 — Pair `model_type` and `dataset_name` based on your data\n\nPrefer the dataset-specific class when your layout matches a supported one — it applies class-specific path conventions, evaluation crops, and (where applicable) occlusion-mask handling. Fall back to `GenericDataset` only for layouts that do not match any registered class.\n\n| Data category | `model_type` | `dataset_name` |\n|---|---|---|\n| Middlebury data | `FoundationStereo` | `Middlebury` |\n| KITTI data | `FoundationStereo` | `Kitti` |\n| ETH3D data | `FoundationStereo` | `Eth3d` |\n| FSD synthetic data | `FoundationStereo` | `FSD` |\n| IsaacReal synthetic data | `FoundationStereo` | `IsaacRealDataset` |\n| Crestereo synthetic data | `FoundationStereo` | `Crestereo` |\n| Other \u002F non-canonical layout | `FoundationStereo` | `GenericDataset` |\n\nValid `dataset_name` values for stereo `data_sources` (case-insensitive): `FSD`, `IsaacRealDataset`, `Crestereo`, `Middlebury`, `Eth3d`, `Kitti`, `GenericDataset`.\n\nThe same `dataset_name` value applies across train and evaluate actions (all of which use 3-column or 4-column annotations with GT disparity). The deploy-side `evaluate` action follows the same rule — see `references\u002Ftao-deploy-foundation-stereo.md`. For inference with 2-column annotations (left + right, no GT), use `dataset_name: GenericDataset` regardless of data layout — the dataset-specific classes (`Middlebury` \u002F `Kitti` \u002F `Eth3d` \u002F `FSD` \u002F `IsaacRealDataset` \u002F `Crestereo`) require 3-column input and reject 2-column annotations at the dataloader level. For inference with 3-column annotations (left + right + GT), the dataset-specific class is fine.\n\n### Step 3 — Write spec yaml from the spec overrides\n\nCopy the action block from `references\u002Fspec-overrides-foundation-stereo.md`. Replace:\n- `model.model_type` from Step 2 (typically `FoundationStereo`)\n- `dataset.\u003C...>.data_sources[*].dataset_name` from Step 2\n- `dataset.\u003C...>.data_sources[*].data_file` with the path from Step 1\n- For deploy-side `evaluate`: enforce `dataset.test_dataset.batch_size: 1` (see `references\u002Ftao-deploy-foundation-stereo.md`).\n\nShape consistency: the `crop_size` in `dataset.test_dataset.augmentation.crop_size` should match `export.input_height` \u002F `input_width` so the trained-model evaluator and the deploy-side TensorRT evaluator operate at the same shape. Note that `crop_size` is decorative on the pyt `evaluate` path but authoritative on the deploy `evaluate` side — see `references\u002Ftroubleshooting-foundation-stereo.md` and `references\u002Ftao-deploy-foundation-stereo.md`.\n\nFresh-install smoke runs are validated at `crop_size: [128, 128]` with `dataset.max_disparity: 128` and `model.max_disparity: 128`. Avoid 112×112 crops and avoid setting `max_disparity` smaller than the square crop side for smoke tests: those combinations can fail inside FoundationStereo with feature-map or loss-mask shape mismatches before a checkpoint is produced.\n\nData source overrides are **mandatory for every action**. Each `data_sources` entry is a dict with two mandatory fields: `data_file` and `dataset_name`. See `references\u002Fspec-overrides-foundation-stereo.md` for the per-action dataset-requirements table, every action's override block, and the `quantize` known-issue note.\n\n### Step 4 — Run\n\nCreate writable home\u002Fcache directories inside the mounted output path before using\n`--user`. Some TAO containers do not have an `\u002Fetc\u002Fpasswd` entry for the host UID,\nand PyTorch \u002F matplotlib need writable cache paths when running as that UID.\n\n```bash\nmkdir -p \u003Coutput_dir>\u002Fhome \\\n         \u003Coutput_dir>\u002F.cache\u002Fmatplotlib \\\n         \u003Coutput_dir>\u002F.cache\u002Ftorchinductor \\\n         \u003Coutput_dir>\u002F.cache\u002Fxdg\n```\n\n```\ndocker run --gpus 'device=0' --shm-size 16G --ipc=host \\\n  --user \"$(id -u):$(id -g)\" \\\n  -e USER=\"$(id -un)\" \\\n  -e LOGNAME=\"$(id -un)\" \\\n  -e HOME=\u003Coutput_dir>\u002Fhome \\\n  -e MPLCONFIGDIR=\u003Coutput_dir>\u002F.cache\u002Fmatplotlib \\\n  -e TORCHINDUCTOR_CACHE_DIR=\u003Coutput_dir>\u002F.cache\u002Ftorchinductor \\\n  -e XDG_CACHE_HOME=\u003Coutput_dir>\u002F.cache\u002Fxdg \\\n  -v \u003Cdata_root>:\u003Cdata_root>:ro \\\n  -v \u003Coutput_dir>:\u003Coutput_dir> \\\n  \u003Ccontainer> \\\n  depth_net \u003Caction> -e \u003Cspec.yaml>\n```\n\nWithout `--user \"$(id -u):$(id -g)\"` the container writes outputs as `nobody:nogroup`, blocking host-side cleanup \u002F retry.\n\n### Step 5 — Verify\n\n- Container exit code 0\n- `status.json` `kpi` block populated\n- For `train`: inspect per-step `train_loss` directly (the entrypoint reports `Execution status: PASS` even when loss is NaN)\n- For `evaluate`: rely on `epe` \u002F `bp1` \u002F `bp2` \u002F `bp3` \u002F `d1` \u002F `rmse` (the evaluator also emits `abs_rel` \u002F `sq_rel` \u002F `rmse_log` which are non-meaningful for stereo — see `references\u002Fparameters-foundation-stereo.md`)\n- For `inference`: artifacts under `results_dir`\n\nFor TAO Deploy TensorRT actions (`gen_trt_engine`, TensorRT `evaluate`, and TensorRT `inference`), read `references\u002Ftao-deploy-foundation-stereo.md` first. Deploy spec templates live in this skill's `references\u002F` folder with the `spec_template_deploy_*.yaml` prefix.\n\n## Training Requirements\n\n- **Monitoring metric:** val\u002Floss\n- **Eval dataset:** optional. Val dataset configured via `dataset.val_dataset.data_sources` (each entry needs `data_file` and `dataset_name`).\n\nSee `references\u002Fspec-overrides-foundation-stereo.md` for the per-action dataset-requirements table and every action's mandatory data-source override block.\n\n## Parameters, Metrics, Multi-GPU, Export\u002FTRT, Hardware\n\nSee `references\u002Fparameters-foundation-stereo.md` for the full Important Parameters list (incl. `model.encoder` `vits` override, `model.max_disparity` default 416, `model.volume_dim` no-op note, `dataset.baseline`, `dataset.focal_x`, `train.precision`, `export.batch_size`), the Evaluation Metrics table, Multi-GPU \u002F Multi-Node launch keys, Export \u002F TRT Defaults (`opset_version`\u002F`on_cpu` pairing, NGC 576×960 settings), and Hardware requirements.\n\n## Error Patterns and Troubleshooting\n\nSee `references\u002Ftroubleshooting-foundation-stereo.md` for disparity overflow, smoke-test shape mismatch, missing pretrained paths, the `encoder` \u002F `dataset_name` struct errors, the `depth_net_stereo: not found` entrypoint note, the pyt-vs-deploy `crop_size` discussion, and the deploy `evaluate` scalar-conversion failure.\n\n## Spec Param \u002F Parent Model Inference\n\nSee `references\u002Fcheckpoint-inference-mappings-foundation-stereo.md` for the checkpoint-resolution rules (`model_epoch_\u003Cepoch>_step_\u003Cstep>.pth`, `dn_model_latest.pth` policy), the absence of parent PyT `gen_trt_engine`, and the full per-action inference-mapping table from `depth_net_stereo.config.json` (including `parent_model` \u002F `parent_job_id` resolution).\n\n## Deployment\n\n- [tao-deploy-foundation-stereo](references\u002Ftao-deploy-foundation-stereo.md)\n","# Depth Net Stereo\n\nStereo depth estimation using FoundationStereo architecture. Predicts disparity maps from stereo image pairs for 3D reconstruction.\n\nUses pretrained Depth Anything v2 and EdgeNeXt encoders. Set `model.stereo_backbone.depth_anything_v2_pretrained_path` and `model.stereo_backbone.edgenext_pretrained_path`.\n\nThe mono and stereo skills both invoke the unified TAO `depth_net` CLI inside the container; the mono\u002Fstereo family is selected via `model.model_type` (e.g., `FoundationStereo`).\n\nPyT actions packaged by this model skill: `train`, `evaluate`, `inference`, `export`, and `quantize`. The PyT `depth_net` entrypoint does not accept a `gen_trt_engine` action in the current TAO image; build TensorRT engines only through the deploy workflow.\n\nFor TAO Deploy TensorRT actions (`gen_trt_engine`, TensorRT `evaluate`, and TensorRT `inference`), read `references\u002Ftao-deploy-foundation-stereo.md` first. The deploy spec template lives in this skill's `references\u002Fspec_template_deploy.yaml`.\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## Workflow\n\n### Prerequisites — data accessibility\n\nYour dataset (left + right images + GT disparity) must be reachable from inside the container:\n- **SDK runner**: place files at the S3 paths the runner resolves (the `S3_TRAIN` \u002F `S3_EVAL` placeholders shown in the spec overrides). The runner handles S3 → container-path mounting transparently.\n- **Direct `docker run`** (e.g. local testing): mount the host dataset root read-only at the same in-container path:\n\n```\ndocker run ... -v \u003Chost_data_root>:\u003Chost_data_root>:ro \u003Ccontainer> ...\n```\n\nThe same accessibility requirement applies to the `\u003Coutput_dir>` written by all actions.\n\n### Step 1 — Annotation file\n\nPer-line annotation file referenced by `data_sources[*].data_file`:\n\n| Columns | Format | Use |\n|---|---|---|\n| 2 | `\u003Cleft> \u003Cright>` | Stereo inference (no GT) |\n| 3 | `\u003Cleft> \u003Cright> \u003Cdisparity>` | Stereo with GT |\n| 4 | `\u003Cleft> \u003Cright> \u003Cdisparity> \u003Cocclusion_mask>` | Stereo with GT and occlusion mask |\n\nIf you already have one, point to it. Otherwise generate via `depth_net convert`:\n\n```\ndepth_net convert -e \u003Cconvert_spec.yaml>\n```\n\n`convert_spec.yaml` template (stereo):\n\n```yaml\nresults_dir: \u003Cdirectory where generated annotation files are written>\ndata_root: \u003Cdirectory whose immediate children are scene folders that contain your image+depth files; convert walks data_root recursively but expects per-scene subdirectories at one level below>\nimage_dir_pattern: [\u003Csubstring matching left image paths>]\nright_dir_pattern: [\u003Csubstring matching right image paths>]\ndepth_dir_pattern: [\u003Csubstring matching GT disparity paths>]\nnocc_dir_pattern: []                 # optional, occlusion mask paths\nimage_extension: '.png'  # always include the leading dot\ndepth_extension: '.png'  # form must match image_extension (the swap is a substring replace)\nnocc_extension: ''\nsplit_ratio: 0.0        # 0.0\u002F1.0 = test-only; 0.8 = 80\u002F20 train+val\n```\n\n`convert` walks `data_root` recursively, selects paths whose path-string contains *all* substrings in `image_dir_pattern` (AND-filter), then derives right \u002F depth \u002F mask paths by replacing `image_dir_pattern[0]` with the corresponding pattern's first element plus extension swap. Inspect your dataset's directory layout and identify the substrings distinguishing left, right, and GT (e.g. `im0` vs `im1` vs `disp0GT` for Middlebury).\n\n### Step 2 — Pair `model_type` and `dataset_name` based on your data\n\nPrefer the dataset-specific class when your layout matches a supported one — it applies class-specific path conventions, evaluation crops, and (where applicable) occlusion-mask handling. Fall back to `GenericDataset` only for layouts that do not match any registered class.\n\n| Data category | `model_type` | `dataset_name` |\n|---|---|---|\n| Middlebury data | `FoundationStereo` | `Middlebury` |\n| KITTI data | `FoundationStereo` | `Kitti` |\n| ETH3D data | `FoundationStereo` | `Eth3d` |\n| FSD synthetic data | `FoundationStereo` | `FSD` |\n| IsaacReal synthetic data | `FoundationStereo` | `IsaacRealDataset` |\n| Crestereo synthetic data | `FoundationStereo` | `Crestereo` |\n| Other \u002F non-canonical layout | `FoundationStereo` | `GenericDataset` |\n\nValid `dataset_name` values for stereo `data_sources` (case-insensitive): `FSD`, `IsaacRealDataset`, `Crestereo`, `Middlebury`, `Eth3d`, `Kitti`, `GenericDataset`.\n\nThe same `dataset_name` value applies across train and evaluate actions (all of which use 3-column or 4-column annotations with GT disparity). The deploy-side `evaluate` action follows the same rule — see `references\u002Ftao-deploy-foundation-stereo.md`. For inference with 2-column annotations (left + right, no GT), use `dataset_name: GenericDataset` regardless of data layout — the dataset-specific classes (`Middlebury` \u002F `Kitti` \u002F `Eth3d` \u002F `FSD` \u002F `IsaacRealDataset` \u002F `Crestereo`) require 3-column input and reject 2-column annotations at the dataloader level. For inference with 3-column annotations (left + right + GT), the dataset-specific class is fine.\n\n### Step 3 — Write spec yaml from the spec overrides\n\nCopy the action block from `references\u002Fspec-overrides-foundation-stereo.md`. Replace:\n- `model.model_type` from Step 2 (typically `FoundationStereo`)\n- `dataset.\u003C...>.data_sources[*].dataset_name` from Step 2\n- `dataset.\u003C...>.data_sources[*].data_file` with the path from Step 1\n- For deploy-side `evaluate`: enforce `dataset.test_dataset.batch_size: 1` (see `references\u002Ftao-deploy-foundation-stereo.md`).\n\nShape consistency: the `crop_size` in `dataset.test_dataset.augmentation.crop_size` should match `export.input_height` \u002F `input_width` so the trained-model evaluator and the deploy-side TensorRT evaluator operate at the same shape. Note that `crop_size` is decorative on the pyt `evaluate` path but authoritative on the deploy `evaluate` side — see `references\u002Ftroubleshooting-foundation-stereo.md` and `references\u002Ftao-deploy-foundation-stereo.md`.\n\nFresh-install smoke runs are validated at `crop_size: [128, 128]` with `dataset.max_disparity: 128` and `model.max_disparity: 128`. Avoid 112×112 crops and avoid setting `max_disparity` smaller than the square crop side for smoke tests: those combinations can fail inside FoundationStereo with feature-map or loss-mask shape mismatches before a checkpoint is produced.\n\nData source overrides are **mandatory for every action**. Each `data_sources` entry is a dict with two mandatory fields: `data_file` and `dataset_name`. See `references\u002Fspec-overrides-foundation-stereo.md` for the per-action dataset-requirements table, every action's override block, and the `quantize` known-issue note.\n\n### Step 4 — Run\n\nCreate writable home\u002Fcache directories inside the mounted output path before using\n`--user`. Some TAO containers do not have an `\u002Fetc\u002Fpasswd` entry for the host UID,\nand PyTorch \u002F matplotlib need writable cache paths when running as that UID.\n\n```bash\nmkdir -p \u003Coutput_dir>\u002Fhome \\\n         \u003Coutput_dir>\u002F.cache\u002Fmatplotlib \\\n         \u003Coutput_dir>\u002F.cache\u002Ftorchinductor \\\n         \u003Coutput_dir>\u002F.cache\u002Fxdg\n```\n\n```\ndocker run --gpus 'device=0' --shm-size 16G --ipc=host \\\n  --user \"$(id -u):$(id -g)\" \\\n  -e USER=\"$(id -un)\" \\\n  -e LOGNAME=\"$(id -un)\" \\\n  -e HOME=\u003Coutput_dir>\u002Fhome \\\n  -e MPLCONFIGDIR=\u003Coutput_dir>\u002F.cache\u002Fmatplotlib \\\n  -e TORCHINDUCTOR_CACHE_DIR=\u003Coutput_dir>\u002F.cache\u002Ftorchinductor \\\n  -e XDG_CACHE_HOME=\u003Coutput_dir>\u002F.cache\u002Fxdg \\\n  -v \u003Cdata_root>:\u003Cdata_root>:ro \\\n  -v \u003Coutput_dir>:\u003Coutput_dir> \\\n  \u003Ccontainer> \\\n  depth_net \u003Caction> -e \u003Cspec.yaml>\n```\n\nWithout `--user \"$(id -u):$(id -g)\"` the container writes outputs as `nobody:nogroup`, blocking host-side cleanup \u002F retry.\n\n### Step 5 — Verify\n\n- Container exit code 0\n- `status.json` `kpi` block populated\n- For `train`: inspect per-step `train_loss` directly (the entrypoint reports `Execution status: PASS` even when loss is NaN)\n- For `evaluate`: rely on `epe` \u002F `bp1` \u002F `bp2` \u002F `bp3` \u002F `d1` \u002F `rmse` (the evaluator also emits `abs_rel` \u002F `sq_rel` \u002F `rmse_log` which are non-meaningful for stereo — see `references\u002Fparameters-foundation-stereo.md`)\n- For `inference`: artifacts under `results_dir`\n\nFor TAO Deploy TensorRT actions (`gen_trt_engine`, TensorRT `evaluate`, and TensorRT `inference`), read `references\u002Ftao-deploy-foundation-stereo.md` first. Deploy spec templates live in this skill's `references\u002F` folder with the `spec_template_deploy_*.yaml` prefix.\n\n## Training Requirements\n\n- **Monitoring metric:** val\u002Floss\n- **Eval dataset:** optional. Val dataset configured via `dataset.val_dataset.data_sources` (each entry needs `data_file` and `dataset_name`).\n\nSee `references\u002Fspec-overrides-foundation-stereo.md` for the per-action dataset-requirements table and every action's mandatory data-source override block.\n\n## Parameters, Metrics, Multi-GPU, Export\u002FTRT, Hardware\n\nSee `references\u002Fparameters-foundation-stereo.md` for the full Important Parameters list (incl. `model.encoder` `vits` override, `model.max_disparity` default 416, `model.volume_dim` no-op note, `dataset.baseline`, `dataset.focal_x`, `train.precision`, `export.batch_size`), the Evaluation Metrics table, Multi-GPU \u002F Multi-Node launch keys, Export \u002F TRT Defaults (`opset_version`\u002F`on_cpu` pairing, NGC 576×960 settings), and Hardware requirements.\n\n## Error Patterns and Troubleshooting\n\nSee `references\u002Ftroubleshooting-foundation-stereo.md` for disparity overflow, smoke-test shape mismatch, missing pretrained paths, the `encoder` \u002F `dataset_name` struct errors, the `depth_net_stereo: not found` entrypoint note, the pyt-vs-deploy `crop_size` discussion, and the deploy `evaluate` scalar-conversion failure.\n\n## Spec Param \u002F Parent Model Inference\n\nSee `references\u002Fcheckpoint-inference-mappings-foundation-stereo.md` for the checkpoint-resolution rules (`model_epoch_\u003Cepoch>_step_\u003Cstep>.pth`, `dn_model_latest.pth` policy), the absence of parent PyT `gen_trt_engine`, and the full per-action inference-mapping table from `depth_net_stereo.config.json` (including `parent_model` \u002F `parent_job_id` resolution).\n\n## Deployment\n\n- [tao-deploy-foundation-stereo](references\u002Ftao-deploy-foundation-stereo.md)\n",{"name":5,"tags":24,"license":15,"metadata":28,"description":7,"allowed-tools":31,"compatibility":32},[25,26,27],"stereo","depth","estimation",{"author":29,"version":30},"NVIDIA Corporation","0.1.0","Read Bash","Requires docker + nvidia-container-toolkit.","main","8859218955d9504cd18b59e7892100718039d3b4","data\u002Fskills\u002Fnvidia-skills\u002Ftao-train-foundation-stereo",1,"https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper","https:\u002F\u002Fgithub.com\u002Fom-scogo","SDE-1 @ Scogo ","2026-07-18 12:22:46",{"source":8,"skillName":5,"commands":42,"prompt":52,"zipAvailable":53},[43,46,49],{"tool":44,"command":45},"npx","npx skills add om-scogo\u002Fskillsh-scraper --skill tao-train-foundation-stereo",{"tool":47,"command":48},"bunx","bunx skills add om-scogo\u002Fskillsh-scraper --skill tao-train-foundation-stereo",{"tool":50,"command":51},"pnpm","pnpm dlx skills add om-scogo\u002Fskillsh-scraper --skill tao-train-foundation-stereo","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fnvidia-skills\u002Ftao-train-foundation-stereo\u002FSkill.md",true,[55],{"path":56,"resourceType":57,"resourceTypeDesc":58,"sizeBytes":59,"sourceHtmlUrl":60},"metadata.json","REFERENCE","参考文档",458,"https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fnvidia-skills\u002Ftao-train-foundation-stereo\u002Fmetadata.json",[62,70,75,82,87,94],{"skillNo":63,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":66,"licenseFileUrl":6,"sourceHtmlUrl":67,"categories":68,"keywords":69,"publishedAt":19},"01M7D6MC9FAAEV4CMNYX8W7JTG","arxiv-search","Search arXiv physics, math, and computer science preprints using natural language queries. Powered by Valyu semantic search.","MIT","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fyorkeccak-scientific-skills\u002Farxiv-search\u002Fskills\u002Farxiv-search\u002FSKILL.md",[],[],{"skillNo":71,"name":64,"nameZh":6,"description":65,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":66,"licenseFileUrl":6,"sourceHtmlUrl":72,"categories":73,"keywords":74,"publishedAt":19},"01M7D6M85CKKN87TA98VCQV4H7","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fyorkeccak-scientific-skills\u002Farxiv-search\u002FSkill.md",[],[],{"skillNo":76,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":66,"licenseFileUrl":6,"sourceHtmlUrl":79,"categories":80,"keywords":81,"publishedAt":19},"01M7D4ZTCS4DR3CE4JY06X84T2","working-with-legacy-code","Safely change and test untested codebases using Feathers' \"Working Effectively with Legacy Code\". Use when the user mentions \"legacy code\", \"no tests\", \"untested codebase\", \"how do I test this\", \"seams\", \"characterization tests\", \"golden master\", \"sprout method\", \"afraid to change this code\", \"monster method\", \"dependency breaking\", or \"inherited a messy codebase\". Also trigger when changing code without tests safely, getting a class under test when constructors, statics, or singletons block it, adding features to tangled modules, or planning incremental test coverage for an old codebase. Covers the legacy-code change algorithm, seams, characterization tests, sprout\u002Fwrap, and dependency-breaking techniques. For refactoring code that already has tests, see refactoring-patterns. For day-to-day code quality, see clean-code.","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fworking-with-legacy-code\u002Fworking-with-legacy-code\u002FSKILL.md",[],[],{"skillNo":83,"name":77,"nameZh":6,"description":78,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":66,"licenseFileUrl":6,"sourceHtmlUrl":84,"categories":85,"keywords":86,"publishedAt":19},"01M7D4ZHZKJD2RSNGAF63V14P4","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fworking-with-legacy-code\u002FSkill.md",[],[],{"skillNo":88,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":66,"licenseFileUrl":6,"sourceHtmlUrl":91,"categories":92,"keywords":93,"publishedAt":19},"01M7D4ZB4CJBFW5JE67CPH2QRV","web-typography","Select, pair, and implement typefaces for web projects. Use when the user mentions \"font pairing\", \"which typeface\", \"line height\", \"responsive typography\", \"web font loading\", \"type hierarchy\", \"variable fonts\", \"FOUT\u002FFOIT\", \"typographic scale\", or \"the text is hard to read\". Also trigger when choosing between system fonts and web fonts, optimizing font-loading performance, or designing readable long-form content. Covers readability evaluation, CSS implementation, and performance optimization. For overall UI design systems, see refactoring-ui. For dramatic typographic experiences, see top-design.","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fweb-typography\u002Fweb-typography\u002FSKILL.md",[],[],{"skillNo":95,"name":89,"nameZh":6,"description":90,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":66,"licenseFileUrl":6,"sourceHtmlUrl":96,"categories":97,"keywords":98,"publishedAt":19},"01M7D4Z4MVYVY3NMJ18BNV6WNA","https:\u002F\u002Fgithub.com\u002Fom-scogo\u002Fskillsh-scraper\u002Fblob\u002Fmain\u002Fdata\u002Fskills\u002Fwondelai-skills\u002Fweb-typography\u002FSkill.md",[],[]]