[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill-nvidia-nv-generate-mr-brain":3,"mdc--b5vd3m-key":31,"related-repo-nvidia-nv-generate-mr-brain":1620,"related-org-nvidia-nv-generate-mr-brain":1724},{"slug":4,"name":4,"fn":5,"description":6,"org":7,"tags":11,"stars":20,"repoUrl":21,"updatedAt":22,"license":23,"forks":24,"topics":25,"repo":26,"sourceUrl":29,"mdContent":30},"nv-generate-mr-brain","generate synthetic brain MRI volumes","Used for generating synthetic brain MRI volumes with NV-Generate-CTMR rflow-mr-brain. Not for production training data.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},"nvidia","NVIDIA","https:\u002F\u002Fpexgzepcugksgbtrxkhf.supabase.co\u002Fstorage\u002Fv1\u002Fobject\u002Fpublic\u002Forg-logos\u002Fnvidia.png",[12,16,17],{"name":13,"slug":14,"type":15},"Healthcare","healthcare","tag",{"name":9,"slug":8,"type":15},{"name":18,"slug":19,"type":15},"Imaging","imaging",2473,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills","2026-07-14T05:29:28.248853","Apache-2.0",281,[],{"repoUrl":21,"stars":20,"forks":24,"topics":27,"description":28},[],"AI agent skills published by NVIDIA","https:\u002F\u002Fgithub.com\u002FNVIDIA\u002Fskills\u002Ftree\u002FHEAD\u002Fskills\u002Fnv-generate-mr-brain","---\nname: nv-generate-mr-brain\ndescription: Used for generating synthetic brain MRI volumes with NV-Generate-CTMR rflow-mr-brain. Not for production training data.\nlicense: Apache-2.0\nallowed-tools: Bash\nmetadata:\n  author: NVIDIA MedTech Team\n  tags:\n    - MedTech\n    - MRI\n    - brain\n---\n\n# NV-Generate-MR-Brain\n\n## Purpose\n- Used for generating synthetic brain MRI volumes with NV-Generate-CTMR rflow-mr-brain. Not for production training data.\n- Use the wrapper exactly as documented; do not replace the upstream entrypoint with a handwritten implementation.\n- Do not write custom inference code for normal runs. The wrapper owns config staging, output paths, and validation.\n- Manifest I\u002FO: inputs are `model_config_override`; outputs are `synthetic_mr_brain_volumes` and `result_json`.\n\n## Instructions\n- Read `skill_manifest.yaml` before changing arguments, side effects, or validation gates.\n- Run `scripts\u002Frun_mr_brain.py` through the documented command below; keep outputs under a caller-provided run directory.\n- If a host agent exposes `run_script`, use `run_script(\"scripts\u002Frun_mr_brain.py\", args=[...])`; otherwise run the Bash\u002FPython command shown below.\n- Emit a single bash code block, and keep the `python -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"` step in that same command — the runtime may be a fresh environment without `nibabel`\u002FMONAI, so dropping the install fails with `ModuleNotFoundError`.\n- Do not add `rm`, `mkdir`, or any cleanup of `--output-dir`; the wrapper creates it. Use a fresh `--output-dir` instead of deleting one.\n- Check the emitted JSON and paired verifier guidance before treating the run as evidence.\n\n## Available Scripts\n| Script | Purpose | Arguments |\n|---|---|---|\n| `scripts\u002Frun_mr_brain.py` | Primary entrypoint declared by skill_manifest.yaml. | `MODEL_CONFIG.json --output-dir OUT_DIR --modality mri_t1 [--random-seed N] [--yes]` |\n\n## Prerequisites\n- Runtime requirements: GPU\u002FCUDA when declared by the manifest; Python packages listed in `runtime.side_effects.pip_packages`.\n- Side effects: writes generated outputs under the caller's `--output-dir`, may cache model assets under `~\u002F.cache\u002Fhuggingface\u002F`, and may contact `https:\u002F\u002Fhuggingface.co` or `https:\u002F\u002Fgithub.com` during setup.\n- Run commands from the repository root unless an existing section below says otherwise.\n\n## Limitations\n- This is a thin wrapper. Inference, sampling, and decoding are delegated entirely to NVIDIA-Medtech\u002FNV-Generate-CTMR's `scripts.diff_model_infer`. Do not modify code under $NV_GENERATE_ROOT or the repo-local fallback at .workbench_data\u002Fupstreams\u002FNV-Generate-CTMR.\n- rflow-mr-brain generates image-only synthetic brain MRI volumes. It does not emit paired segmentation masks.\n- Output volumes are synthetic. They are not safe as training data for production medtech models without independent quality review.\n- Not for clinical deployment, clinical interpretation, autonomous diagnosis, regulatory submission.\n\n## Troubleshooting\n| Error | Cause | Fix |\n|---|---|---|\n| Missing dependency or import error | Runtime package drift from `skill_manifest.yaml`. | Install the packages declared in the manifest or use the documented setup command. |\n| Empty or schema-invalid output | Wrong input path, unsupported modality, or upstream failure. | Re-run with a known fixture and inspect the wrapper JSON plus stderr. |\n| Validation gate failure | Output violated a declared engineering invariant. | Keep the failed evidence pack and use the gate message to repair inputs or wrapper code. |\n\nWraps the upstream\n[`NVIDIA-Medtech\u002FNV-Generate-CTMR`](https:\u002F\u002Fgithub.com\u002FNVIDIA-Medtech\u002FNV-Generate-CTMR\u002Ftree\u002F61c4ec709b84cad468852243c48e250bec732074)\nMR brain image-only generation workflow. The wrapper does not reimplement\ndiffusion sampling or autoencoder decoding. It stages config overrides, runs\nthe documented `python -m scripts.diff_model_infer` command for\n`rflow-mr-brain`, then summarizes the generated NIfTI volume.\n\n\n## Exact Runnable Surface\n\nFor user run commands, use this repo-root wrapper path exactly:\n\n```bash\nexport NV_GENERATE_ROOT=\"${NV_GENERATE_ROOT:-$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7}\" && \\\npython -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\" && \\\npython skills\u002Fnv-generate-mr-brain\u002Fscripts\u002Frun_mr_brain.py PATH_TO_MR_BRAIN_CONFIG.json --output-dir OUT_DIR --modality mri_t1 --random-seed 1234\n```\n\nDo not invent `generate.sh`, `infer.py`, `Medical AI Skills run`, or `python -m nv_generate_mr_brain` commands. `PATH_TO_MR_BRAIN_CONFIG.json` must be the user's supplied request path.\n\n## Preconditions\n\nIf `NV_GENERATE_ROOT` already names a local checkout, the wrapper uses it and\nrecords its current commit in the result. Otherwise, create the recommended\npinned default checkout once:\n\n```bash\nif [ -z \"${NV_GENERATE_ROOT:-}\" ]; then\n  export NV_GENERATE_COMMIT=61c4ec709b84cad468852243c48e250bec732074\n  export NV_GENERATE_ROOT=\"$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7\"\n  if [ ! -d \"$NV_GENERATE_ROOT\u002F.git\" ]; then\n    git clone https:\u002F\u002Fgithub.com\u002FNVIDIA-Medtech\u002FNV-Generate-CTMR.git \"$NV_GENERATE_ROOT\"\n    git -C \"$NV_GENERATE_ROOT\" checkout --detach \"$NV_GENERATE_COMMIT\"\n  fi\nfi\npip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"\n```\n\nDownload the MR-brain weights:\n\n```bash\ncd \"$NV_GENERATE_ROOT\"\npython -m scripts.download_model_data --version rflow-mr-brain --root_dir .\u002F --model_only\n```\n\nRuntime needs an NVIDIA GPU with at least 16 GB VRAM. There is no CPU\nfallback in the upstream path.\n\nThe wrapper also searches `.workbench_data\u002Fupstreams\u002FNV-Generate-CTMR` if\n`NV_GENERATE_ROOT` is unset or does not have the required upstream layout.\n\nFor agent-generated user run commands, use the command in Usage. Do not prepend\nclone or model-download setup steps when the repo-local\nupstream cache already exists. In a fresh Python environment, still include\n`pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"` before the wrapper unless\nthe active environment has already proven those imports are available; cached\nweights do not imply cached Python packages. If setup requires `cd \"$NV_GENERATE_ROOT\"`, return to the Medical AI Skills repo before invoking\n`skills\u002Fnv-generate-mr-brain\u002Fscripts\u002Frun_mr_brain.py`.\n\n## Usage\n\n```bash\nexport NV_GENERATE_ROOT=\"${NV_GENERATE_ROOT:-$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7}\" && \\\npython -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\" && \\\npython skills\u002Fnv-generate-mr-brain\u002Fscripts\u002Frun_mr_brain.py \\\n  PATH_TO_MR_BRAIN_CONFIG.json \\\n  --output-dir runs\u002Fnv_generate_mr_brain_demo \\\n  --modality mri_t1 \\\n  --random-seed 1234\n```\n\nReplace `PATH_TO_MR_BRAIN_CONFIG.json` with the user's actual request\u002Fconfig\npath. Do not copy the fixture path from this document unless the user\nexplicitly asked to run that fixture. If the user says \"the request is at\n`runs\u002F...\u002Fdefault_mri_t1.json`\", that exact path is the first positional\nargument to `scripts\u002Frun_mr_brain.py`.\n\nSupported MR-brain modality names are `mri`, `mri_t1`, `mri_t2`,\n`mri_flair`, `mri_swi`, `mri_t1_skull_stripped`,\n`mri_t2_skull_stripped`, `mri_flair_skull_stripped`, and\n`mri_swi_skull_stripped`. These map to the upstream\n`configs\u002Fmodality_mapping.json` IDs documented in the README.\nFor FOV and setup details, see `references\u002Ffov-and-downloads.md`.\n\nThe fixture argument is a small JSON override for\n`configs\u002Fconfig_maisi_diff_model_rflow-mr-brain.json`. Pass `default` to use\nthe upstream defaults plus the CLI modality and random seed. Common override\nkeys are `dim`, `spacing`, `num_inference_steps`, `cfg_guidance_scale`, and\n`modality`.\n\nEach run records the staged config, model inventory, upstream command, output\ngeometry, spacing, affine, intensity range, and non-constant \u002F finite-data\nchecks. Output volumes are synthetic and are not safe as production training\ndata without independent review.\n\nNot for clinical interpretation, production deployment, autonomous diagnosis,\nor regulatory submission.\n",{"data":32,"body":40},{"name":4,"description":6,"license":23,"allowed-tools":33,"metadata":34},"Bash",{"author":35,"tags":36},"NVIDIA MedTech Team",[37,38,39],"MedTech","MRI","brain",{"type":41,"children":42},"root",[43,51,58,107,113,232,238,295,301,357,363,394,400,487,524,530,535,754,797,803,815,1098,1103,1172,1177,1197,1225,1231,1436,1462,1548,1604,1609,1614],{"type":44,"tag":45,"props":46,"children":47},"element","h1",{"id":4},[48],{"type":49,"value":50},"text","NV-Generate-MR-Brain",{"type":44,"tag":52,"props":53,"children":55},"h2",{"id":54},"purpose",[56],{"type":49,"value":57},"Purpose",{"type":44,"tag":59,"props":60,"children":61},"ul",{},[62,67,72,77],{"type":44,"tag":63,"props":64,"children":65},"li",{},[66],{"type":49,"value":6},{"type":44,"tag":63,"props":68,"children":69},{},[70],{"type":49,"value":71},"Use the wrapper exactly as documented; do not replace the upstream entrypoint with a handwritten implementation.",{"type":44,"tag":63,"props":73,"children":74},{},[75],{"type":49,"value":76},"Do not write custom inference code for normal runs. The wrapper owns config staging, output paths, and validation.",{"type":44,"tag":63,"props":78,"children":79},{},[80,82,89,91,97,99,105],{"type":49,"value":81},"Manifest I\u002FO: inputs are ",{"type":44,"tag":83,"props":84,"children":86},"code",{"className":85},[],[87],{"type":49,"value":88},"model_config_override",{"type":49,"value":90},"; outputs are ",{"type":44,"tag":83,"props":92,"children":94},{"className":93},[],[95],{"type":49,"value":96},"synthetic_mr_brain_volumes",{"type":49,"value":98}," and ",{"type":44,"tag":83,"props":100,"children":102},{"className":101},[],[103],{"type":49,"value":104},"result_json",{"type":49,"value":106},".",{"type":44,"tag":52,"props":108,"children":110},{"id":109},"instructions",[111],{"type":49,"value":112},"Instructions",{"type":44,"tag":59,"props":114,"children":115},{},[116,129,142,163,191,227],{"type":44,"tag":63,"props":117,"children":118},{},[119,121,127],{"type":49,"value":120},"Read ",{"type":44,"tag":83,"props":122,"children":124},{"className":123},[],[125],{"type":49,"value":126},"skill_manifest.yaml",{"type":49,"value":128}," before changing arguments, side effects, or validation gates.",{"type":44,"tag":63,"props":130,"children":131},{},[132,134,140],{"type":49,"value":133},"Run ",{"type":44,"tag":83,"props":135,"children":137},{"className":136},[],[138],{"type":49,"value":139},"scripts\u002Frun_mr_brain.py",{"type":49,"value":141}," through the documented command below; keep outputs under a caller-provided run directory.",{"type":44,"tag":63,"props":143,"children":144},{},[145,147,153,155,161],{"type":49,"value":146},"If a host agent exposes ",{"type":44,"tag":83,"props":148,"children":150},{"className":149},[],[151],{"type":49,"value":152},"run_script",{"type":49,"value":154},", use ",{"type":44,"tag":83,"props":156,"children":158},{"className":157},[],[159],{"type":49,"value":160},"run_script(\"scripts\u002Frun_mr_brain.py\", args=[...])",{"type":49,"value":162},"; otherwise run the Bash\u002FPython command shown below.",{"type":44,"tag":63,"props":164,"children":165},{},[166,168,174,176,182,184,190],{"type":49,"value":167},"Emit a single bash code block, and keep the ",{"type":44,"tag":83,"props":169,"children":171},{"className":170},[],[172],{"type":49,"value":173},"python -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"",{"type":49,"value":175}," step in that same command — the runtime may be a fresh environment without ",{"type":44,"tag":83,"props":177,"children":179},{"className":178},[],[180],{"type":49,"value":181},"nibabel",{"type":49,"value":183},"\u002FMONAI, so dropping the install fails with ",{"type":44,"tag":83,"props":185,"children":187},{"className":186},[],[188],{"type":49,"value":189},"ModuleNotFoundError",{"type":49,"value":106},{"type":44,"tag":63,"props":192,"children":193},{},[194,196,202,204,210,212,218,220,225],{"type":49,"value":195},"Do not add ",{"type":44,"tag":83,"props":197,"children":199},{"className":198},[],[200],{"type":49,"value":201},"rm",{"type":49,"value":203},", ",{"type":44,"tag":83,"props":205,"children":207},{"className":206},[],[208],{"type":49,"value":209},"mkdir",{"type":49,"value":211},", or any cleanup of ",{"type":44,"tag":83,"props":213,"children":215},{"className":214},[],[216],{"type":49,"value":217},"--output-dir",{"type":49,"value":219},"; the wrapper creates it. Use a fresh ",{"type":44,"tag":83,"props":221,"children":223},{"className":222},[],[224],{"type":49,"value":217},{"type":49,"value":226}," instead of deleting one.",{"type":44,"tag":63,"props":228,"children":229},{},[230],{"type":49,"value":231},"Check the emitted JSON and paired verifier guidance before treating the run as evidence.",{"type":44,"tag":52,"props":233,"children":235},{"id":234},"available-scripts",[236],{"type":49,"value":237},"Available Scripts",{"type":44,"tag":239,"props":240,"children":241},"table",{},[242,265],{"type":44,"tag":243,"props":244,"children":245},"thead",{},[246],{"type":44,"tag":247,"props":248,"children":249},"tr",{},[250,256,260],{"type":44,"tag":251,"props":252,"children":253},"th",{},[254],{"type":49,"value":255},"Script",{"type":44,"tag":251,"props":257,"children":258},{},[259],{"type":49,"value":57},{"type":44,"tag":251,"props":261,"children":262},{},[263],{"type":49,"value":264},"Arguments",{"type":44,"tag":266,"props":267,"children":268},"tbody",{},[269],{"type":44,"tag":247,"props":270,"children":271},{},[272,281,286],{"type":44,"tag":273,"props":274,"children":275},"td",{},[276],{"type":44,"tag":83,"props":277,"children":279},{"className":278},[],[280],{"type":49,"value":139},{"type":44,"tag":273,"props":282,"children":283},{},[284],{"type":49,"value":285},"Primary entrypoint declared by skill_manifest.yaml.",{"type":44,"tag":273,"props":287,"children":288},{},[289],{"type":44,"tag":83,"props":290,"children":292},{"className":291},[],[293],{"type":49,"value":294},"MODEL_CONFIG.json --output-dir OUT_DIR --modality mri_t1 [--random-seed N] [--yes]",{"type":44,"tag":52,"props":296,"children":298},{"id":297},"prerequisites",[299],{"type":49,"value":300},"Prerequisites",{"type":44,"tag":59,"props":302,"children":303},{},[304,316,352],{"type":44,"tag":63,"props":305,"children":306},{},[307,309,315],{"type":49,"value":308},"Runtime requirements: GPU\u002FCUDA when declared by the manifest; Python packages listed in ",{"type":44,"tag":83,"props":310,"children":312},{"className":311},[],[313],{"type":49,"value":314},"runtime.side_effects.pip_packages",{"type":49,"value":106},{"type":44,"tag":63,"props":317,"children":318},{},[319,321,326,328,334,336,342,344,350],{"type":49,"value":320},"Side effects: writes generated outputs under the caller's ",{"type":44,"tag":83,"props":322,"children":324},{"className":323},[],[325],{"type":49,"value":217},{"type":49,"value":327},", may cache model assets under ",{"type":44,"tag":83,"props":329,"children":331},{"className":330},[],[332],{"type":49,"value":333},"~\u002F.cache\u002Fhuggingface\u002F",{"type":49,"value":335},", and may contact ",{"type":44,"tag":83,"props":337,"children":339},{"className":338},[],[340],{"type":49,"value":341},"https:\u002F\u002Fhuggingface.co",{"type":49,"value":343}," or ",{"type":44,"tag":83,"props":345,"children":347},{"className":346},[],[348],{"type":49,"value":349},"https:\u002F\u002Fgithub.com",{"type":49,"value":351}," during setup.",{"type":44,"tag":63,"props":353,"children":354},{},[355],{"type":49,"value":356},"Run commands from the repository root unless an existing section below says otherwise.",{"type":44,"tag":52,"props":358,"children":360},{"id":359},"limitations",[361],{"type":49,"value":362},"Limitations",{"type":44,"tag":59,"props":364,"children":365},{},[366,379,384,389],{"type":44,"tag":63,"props":367,"children":368},{},[369,371,377],{"type":49,"value":370},"This is a thin wrapper. Inference, sampling, and decoding are delegated entirely to NVIDIA-Medtech\u002FNV-Generate-CTMR's ",{"type":44,"tag":83,"props":372,"children":374},{"className":373},[],[375],{"type":49,"value":376},"scripts.diff_model_infer",{"type":49,"value":378},". Do not modify code under $NV_GENERATE_ROOT or the repo-local fallback at .workbench_data\u002Fupstreams\u002FNV-Generate-CTMR.",{"type":44,"tag":63,"props":380,"children":381},{},[382],{"type":49,"value":383},"rflow-mr-brain generates image-only synthetic brain MRI volumes. It does not emit paired segmentation masks.",{"type":44,"tag":63,"props":385,"children":386},{},[387],{"type":49,"value":388},"Output volumes are synthetic. They are not safe as training data for production medtech models without independent quality review.",{"type":44,"tag":63,"props":390,"children":391},{},[392],{"type":49,"value":393},"Not for clinical deployment, clinical interpretation, autonomous diagnosis, regulatory submission.",{"type":44,"tag":52,"props":395,"children":397},{"id":396},"troubleshooting",[398],{"type":49,"value":399},"Troubleshooting",{"type":44,"tag":239,"props":401,"children":402},{},[403,424],{"type":44,"tag":243,"props":404,"children":405},{},[406],{"type":44,"tag":247,"props":407,"children":408},{},[409,414,419],{"type":44,"tag":251,"props":410,"children":411},{},[412],{"type":49,"value":413},"Error",{"type":44,"tag":251,"props":415,"children":416},{},[417],{"type":49,"value":418},"Cause",{"type":44,"tag":251,"props":420,"children":421},{},[422],{"type":49,"value":423},"Fix",{"type":44,"tag":266,"props":425,"children":426},{},[427,451,469],{"type":44,"tag":247,"props":428,"children":429},{},[430,435,446],{"type":44,"tag":273,"props":431,"children":432},{},[433],{"type":49,"value":434},"Missing dependency or import error",{"type":44,"tag":273,"props":436,"children":437},{},[438,440,445],{"type":49,"value":439},"Runtime package drift from ",{"type":44,"tag":83,"props":441,"children":443},{"className":442},[],[444],{"type":49,"value":126},{"type":49,"value":106},{"type":44,"tag":273,"props":447,"children":448},{},[449],{"type":49,"value":450},"Install the packages declared in the manifest or use the documented setup command.",{"type":44,"tag":247,"props":452,"children":453},{},[454,459,464],{"type":44,"tag":273,"props":455,"children":456},{},[457],{"type":49,"value":458},"Empty or schema-invalid output",{"type":44,"tag":273,"props":460,"children":461},{},[462],{"type":49,"value":463},"Wrong input path, unsupported modality, or upstream failure.",{"type":44,"tag":273,"props":465,"children":466},{},[467],{"type":49,"value":468},"Re-run with a known fixture and inspect the wrapper JSON plus stderr.",{"type":44,"tag":247,"props":470,"children":471},{},[472,477,482],{"type":44,"tag":273,"props":473,"children":474},{},[475],{"type":49,"value":476},"Validation gate failure",{"type":44,"tag":273,"props":478,"children":479},{},[480],{"type":49,"value":481},"Output violated a declared engineering invariant.",{"type":44,"tag":273,"props":483,"children":484},{},[485],{"type":49,"value":486},"Keep the failed evidence pack and use the gate message to repair inputs or wrapper code.",{"type":44,"tag":488,"props":489,"children":490},"p",{},[491,493,506,508,514,516,522],{"type":49,"value":492},"Wraps the upstream\n",{"type":44,"tag":494,"props":495,"children":499},"a",{"href":496,"rel":497},"https:\u002F\u002Fgithub.com\u002FNVIDIA-Medtech\u002FNV-Generate-CTMR\u002Ftree\u002F61c4ec709b84cad468852243c48e250bec732074",[498],"nofollow",[500],{"type":44,"tag":83,"props":501,"children":503},{"className":502},[],[504],{"type":49,"value":505},"NVIDIA-Medtech\u002FNV-Generate-CTMR",{"type":49,"value":507},"\nMR brain image-only generation workflow. The wrapper does not reimplement\ndiffusion sampling or autoencoder decoding. It stages config overrides, runs\nthe documented ",{"type":44,"tag":83,"props":509,"children":511},{"className":510},[],[512],{"type":49,"value":513},"python -m scripts.diff_model_infer",{"type":49,"value":515}," command for\n",{"type":44,"tag":83,"props":517,"children":519},{"className":518},[],[520],{"type":49,"value":521},"rflow-mr-brain",{"type":49,"value":523},", then summarizes the generated NIfTI volume.",{"type":44,"tag":52,"props":525,"children":527},{"id":526},"exact-runnable-surface",[528],{"type":49,"value":529},"Exact Runnable Surface",{"type":44,"tag":488,"props":531,"children":532},{},[533],{"type":49,"value":534},"For user run commands, use this repo-root wrapper path exactly:",{"type":44,"tag":536,"props":537,"children":542},"pre",{"className":538,"code":539,"language":540,"meta":541,"style":541},"language-bash shiki shiki-themes material-theme-lighter material-theme material-theme-palenight","export NV_GENERATE_ROOT=\"${NV_GENERATE_ROOT:-$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7}\" && \\\npython -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\" && \\\npython skills\u002Fnv-generate-mr-brain\u002Fscripts\u002Frun_mr_brain.py PATH_TO_MR_BRAIN_CONFIG.json --output-dir OUT_DIR --modality mri_t1 --random-seed 1234\n","bash","",[543],{"type":44,"tag":83,"props":544,"children":545},{"__ignoreMap":541},[546,647,705],{"type":44,"tag":547,"props":548,"children":551},"span",{"class":549,"line":550},"line",1,[552,558,564,570,575,580,585,590,595,600,605,609,614,618,623,627,632,637,642],{"type":44,"tag":547,"props":553,"children":555},{"style":554},"--shiki-light:#9C3EDA;--shiki-default:#C792EA;--shiki-dark:#C792EA",[556],{"type":49,"value":557},"export",{"type":44,"tag":547,"props":559,"children":561},{"style":560},"--shiki-light:#90A4AE;--shiki-default:#EEFFFF;--shiki-dark:#BABED8",[562],{"type":49,"value":563}," NV_GENERATE_ROOT",{"type":44,"tag":547,"props":565,"children":567},{"style":566},"--shiki-light:#39ADB5;--shiki-default:#89DDFF;--shiki-dark:#89DDFF",[568],{"type":49,"value":569},"=",{"type":44,"tag":547,"props":571,"children":572},{"style":566},[573],{"type":49,"value":574},"\"${",{"type":44,"tag":547,"props":576,"children":577},{"style":560},[578],{"type":49,"value":579},"NV_GENERATE_ROOT",{"type":44,"tag":547,"props":581,"children":582},{"style":566},[583],{"type":49,"value":584},":-",{"type":44,"tag":547,"props":586,"children":587},{"style":560},[588],{"type":49,"value":589},"$HOME",{"type":44,"tag":547,"props":591,"children":592},{"style":566},[593],{"type":49,"value":594},"\u002F",{"type":44,"tag":547,"props":596,"children":598},{"style":597},"--shiki-light:#91B859;--shiki-default:#C3E88D;--shiki-dark:#C3E88D",[599],{"type":49,"value":106},{"type":44,"tag":547,"props":601,"children":602},{"style":560},[603],{"type":49,"value":604},"cache",{"type":44,"tag":547,"props":606,"children":607},{"style":566},[608],{"type":49,"value":594},{"type":44,"tag":547,"props":610,"children":611},{"style":560},[612],{"type":49,"value":613},"nvidia-skills",{"type":44,"tag":547,"props":615,"children":616},{"style":566},[617],{"type":49,"value":594},{"type":44,"tag":547,"props":619,"children":620},{"style":560},[621],{"type":49,"value":622},"upstreams",{"type":44,"tag":547,"props":624,"children":625},{"style":566},[626],{"type":49,"value":594},{"type":44,"tag":547,"props":628,"children":629},{"style":560},[630],{"type":49,"value":631},"NV-Generate-CTMR-61c4ec7",{"type":44,"tag":547,"props":633,"children":634},{"style":566},[635],{"type":49,"value":636},"}\"",{"type":44,"tag":547,"props":638,"children":639},{"style":566},[640],{"type":49,"value":641}," &&",{"type":44,"tag":547,"props":643,"children":644},{"style":560},[645],{"type":49,"value":646}," \\\n",{"type":44,"tag":547,"props":648,"children":650},{"class":549,"line":649},2,[651,657,662,667,672,677,682,687,692,697,701],{"type":44,"tag":547,"props":652,"children":654},{"style":653},"--shiki-light:#E2931D;--shiki-default:#FFCB6B;--shiki-dark:#FFCB6B",[655],{"type":49,"value":656},"python",{"type":44,"tag":547,"props":658,"children":659},{"style":597},[660],{"type":49,"value":661}," -m",{"type":44,"tag":547,"props":663,"children":664},{"style":597},[665],{"type":49,"value":666}," pip",{"type":44,"tag":547,"props":668,"children":669},{"style":597},[670],{"type":49,"value":671}," install",{"type":44,"tag":547,"props":673,"children":674},{"style":597},[675],{"type":49,"value":676}," -r",{"type":44,"tag":547,"props":678,"children":679},{"style":566},[680],{"type":49,"value":681}," \"",{"type":44,"tag":547,"props":683,"children":684},{"style":560},[685],{"type":49,"value":686},"$NV_GENERATE_ROOT",{"type":44,"tag":547,"props":688,"children":689},{"style":597},[690],{"type":49,"value":691},"\u002Frequirements.txt",{"type":44,"tag":547,"props":693,"children":694},{"style":566},[695],{"type":49,"value":696},"\"",{"type":44,"tag":547,"props":698,"children":699},{"style":566},[700],{"type":49,"value":641},{"type":44,"tag":547,"props":702,"children":703},{"style":560},[704],{"type":49,"value":646},{"type":44,"tag":547,"props":706,"children":708},{"class":549,"line":707},3,[709,713,718,723,728,733,738,743,748],{"type":44,"tag":547,"props":710,"children":711},{"style":653},[712],{"type":49,"value":656},{"type":44,"tag":547,"props":714,"children":715},{"style":597},[716],{"type":49,"value":717}," skills\u002Fnv-generate-mr-brain\u002Fscripts\u002Frun_mr_brain.py",{"type":44,"tag":547,"props":719,"children":720},{"style":597},[721],{"type":49,"value":722}," PATH_TO_MR_BRAIN_CONFIG.json",{"type":44,"tag":547,"props":724,"children":725},{"style":597},[726],{"type":49,"value":727}," --output-dir",{"type":44,"tag":547,"props":729,"children":730},{"style":597},[731],{"type":49,"value":732}," OUT_DIR",{"type":44,"tag":547,"props":734,"children":735},{"style":597},[736],{"type":49,"value":737}," --modality",{"type":44,"tag":547,"props":739,"children":740},{"style":597},[741],{"type":49,"value":742}," mri_t1",{"type":44,"tag":547,"props":744,"children":745},{"style":597},[746],{"type":49,"value":747}," --random-seed",{"type":44,"tag":547,"props":749,"children":751},{"style":750},"--shiki-light:#F76D47;--shiki-default:#F78C6C;--shiki-dark:#F78C6C",[752],{"type":49,"value":753}," 1234\n",{"type":44,"tag":488,"props":755,"children":756},{},[757,759,765,766,772,773,779,781,787,789,795],{"type":49,"value":758},"Do not invent ",{"type":44,"tag":83,"props":760,"children":762},{"className":761},[],[763],{"type":49,"value":764},"generate.sh",{"type":49,"value":203},{"type":44,"tag":83,"props":767,"children":769},{"className":768},[],[770],{"type":49,"value":771},"infer.py",{"type":49,"value":203},{"type":44,"tag":83,"props":774,"children":776},{"className":775},[],[777],{"type":49,"value":778},"Medical AI Skills run",{"type":49,"value":780},", or ",{"type":44,"tag":83,"props":782,"children":784},{"className":783},[],[785],{"type":49,"value":786},"python -m nv_generate_mr_brain",{"type":49,"value":788}," commands. ",{"type":44,"tag":83,"props":790,"children":792},{"className":791},[],[793],{"type":49,"value":794},"PATH_TO_MR_BRAIN_CONFIG.json",{"type":49,"value":796}," must be the user's supplied request path.",{"type":44,"tag":52,"props":798,"children":800},{"id":799},"preconditions",[801],{"type":49,"value":802},"Preconditions",{"type":44,"tag":488,"props":804,"children":805},{},[806,808,813],{"type":49,"value":807},"If ",{"type":44,"tag":83,"props":809,"children":811},{"className":810},[],[812],{"type":49,"value":579},{"type":49,"value":814}," already names a local checkout, the wrapper uses it and\nrecords its current commit in the result. Otherwise, create the recommended\npinned default checkout once:",{"type":44,"tag":536,"props":816,"children":818},{"className":538,"code":817,"language":540,"meta":541,"style":541},"if [ -z \"${NV_GENERATE_ROOT:-}\" ]; then\n  export NV_GENERATE_COMMIT=61c4ec709b84cad468852243c48e250bec732074\n  export NV_GENERATE_ROOT=\"$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7\"\n  if [ ! -d \"$NV_GENERATE_ROOT\u002F.git\" ]; then\n    git clone https:\u002F\u002Fgithub.com\u002FNVIDIA-Medtech\u002FNV-Generate-CTMR.git \"$NV_GENERATE_ROOT\"\n    git -C \"$NV_GENERATE_ROOT\" checkout --detach \"$NV_GENERATE_COMMIT\"\n  fi\nfi\npip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"\n",[819],{"type":44,"tag":83,"props":820,"children":821},{"__ignoreMap":541},[822,865,887,920,968,999,1047,1056,1065],{"type":44,"tag":547,"props":823,"children":824},{"class":549,"line":550},[825,831,836,841,846,850,855,860],{"type":44,"tag":547,"props":826,"children":828},{"style":827},"--shiki-light:#39ADB5;--shiki-light-font-style:italic;--shiki-default:#89DDFF;--shiki-default-font-style:italic;--shiki-dark:#89DDFF;--shiki-dark-font-style:italic",[829],{"type":49,"value":830},"if",{"type":44,"tag":547,"props":832,"children":833},{"style":566},[834],{"type":49,"value":835}," [",{"type":44,"tag":547,"props":837,"children":838},{"style":566},[839],{"type":49,"value":840}," -z",{"type":44,"tag":547,"props":842,"children":843},{"style":566},[844],{"type":49,"value":845}," \"${",{"type":44,"tag":547,"props":847,"children":848},{"style":560},[849],{"type":49,"value":579},{"type":44,"tag":547,"props":851,"children":852},{"style":566},[853],{"type":49,"value":854},":-}\"",{"type":44,"tag":547,"props":856,"children":857},{"style":566},[858],{"type":49,"value":859}," ];",{"type":44,"tag":547,"props":861,"children":862},{"style":827},[863],{"type":49,"value":864}," then\n",{"type":44,"tag":547,"props":866,"children":867},{"class":549,"line":649},[868,873,878,882],{"type":44,"tag":547,"props":869,"children":870},{"style":554},[871],{"type":49,"value":872},"  export",{"type":44,"tag":547,"props":874,"children":875},{"style":560},[876],{"type":49,"value":877}," NV_GENERATE_COMMIT",{"type":44,"tag":547,"props":879,"children":880},{"style":566},[881],{"type":49,"value":569},{"type":44,"tag":547,"props":883,"children":884},{"style":560},[885],{"type":49,"value":886},"61c4ec709b84cad468852243c48e250bec732074\n",{"type":44,"tag":547,"props":888,"children":889},{"class":549,"line":707},[890,894,898,902,906,910,915],{"type":44,"tag":547,"props":891,"children":892},{"style":554},[893],{"type":49,"value":872},{"type":44,"tag":547,"props":895,"children":896},{"style":560},[897],{"type":49,"value":563},{"type":44,"tag":547,"props":899,"children":900},{"style":566},[901],{"type":49,"value":569},{"type":44,"tag":547,"props":903,"children":904},{"style":566},[905],{"type":49,"value":696},{"type":44,"tag":547,"props":907,"children":908},{"style":560},[909],{"type":49,"value":589},{"type":44,"tag":547,"props":911,"children":912},{"style":597},[913],{"type":49,"value":914},"\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7",{"type":44,"tag":547,"props":916,"children":917},{"style":566},[918],{"type":49,"value":919},"\"\n",{"type":44,"tag":547,"props":921,"children":923},{"class":549,"line":922},4,[924,929,933,938,943,947,951,956,960,964],{"type":44,"tag":547,"props":925,"children":926},{"style":827},[927],{"type":49,"value":928},"  if",{"type":44,"tag":547,"props":930,"children":931},{"style":566},[932],{"type":49,"value":835},{"type":44,"tag":547,"props":934,"children":935},{"style":566},[936],{"type":49,"value":937}," !",{"type":44,"tag":547,"props":939,"children":940},{"style":566},[941],{"type":49,"value":942}," -d",{"type":44,"tag":547,"props":944,"children":945},{"style":566},[946],{"type":49,"value":681},{"type":44,"tag":547,"props":948,"children":949},{"style":560},[950],{"type":49,"value":686},{"type":44,"tag":547,"props":952,"children":953},{"style":597},[954],{"type":49,"value":955},"\u002F.git",{"type":44,"tag":547,"props":957,"children":958},{"style":566},[959],{"type":49,"value":696},{"type":44,"tag":547,"props":961,"children":962},{"style":566},[963],{"type":49,"value":859},{"type":44,"tag":547,"props":965,"children":966},{"style":827},[967],{"type":49,"value":864},{"type":44,"tag":547,"props":969,"children":971},{"class":549,"line":970},5,[972,977,982,987,991,995],{"type":44,"tag":547,"props":973,"children":974},{"style":653},[975],{"type":49,"value":976},"    git",{"type":44,"tag":547,"props":978,"children":979},{"style":597},[980],{"type":49,"value":981}," clone",{"type":44,"tag":547,"props":983,"children":984},{"style":597},[985],{"type":49,"value":986}," https:\u002F\u002Fgithub.com\u002FNVIDIA-Medtech\u002FNV-Generate-CTMR.git",{"type":44,"tag":547,"props":988,"children":989},{"style":566},[990],{"type":49,"value":681},{"type":44,"tag":547,"props":992,"children":993},{"style":560},[994],{"type":49,"value":686},{"type":44,"tag":547,"props":996,"children":997},{"style":566},[998],{"type":49,"value":919},{"type":44,"tag":547,"props":1000,"children":1002},{"class":549,"line":1001},6,[1003,1007,1012,1016,1020,1024,1029,1034,1038,1043],{"type":44,"tag":547,"props":1004,"children":1005},{"style":653},[1006],{"type":49,"value":976},{"type":44,"tag":547,"props":1008,"children":1009},{"style":597},[1010],{"type":49,"value":1011}," -C",{"type":44,"tag":547,"props":1013,"children":1014},{"style":566},[1015],{"type":49,"value":681},{"type":44,"tag":547,"props":1017,"children":1018},{"style":560},[1019],{"type":49,"value":686},{"type":44,"tag":547,"props":1021,"children":1022},{"style":566},[1023],{"type":49,"value":696},{"type":44,"tag":547,"props":1025,"children":1026},{"style":597},[1027],{"type":49,"value":1028}," checkout",{"type":44,"tag":547,"props":1030,"children":1031},{"style":597},[1032],{"type":49,"value":1033}," --detach",{"type":44,"tag":547,"props":1035,"children":1036},{"style":566},[1037],{"type":49,"value":681},{"type":44,"tag":547,"props":1039,"children":1040},{"style":560},[1041],{"type":49,"value":1042},"$NV_GENERATE_COMMIT",{"type":44,"tag":547,"props":1044,"children":1045},{"style":566},[1046],{"type":49,"value":919},{"type":44,"tag":547,"props":1048,"children":1050},{"class":549,"line":1049},7,[1051],{"type":44,"tag":547,"props":1052,"children":1053},{"style":827},[1054],{"type":49,"value":1055},"  fi\n",{"type":44,"tag":547,"props":1057,"children":1059},{"class":549,"line":1058},8,[1060],{"type":44,"tag":547,"props":1061,"children":1062},{"style":827},[1063],{"type":49,"value":1064},"fi\n",{"type":44,"tag":547,"props":1066,"children":1068},{"class":549,"line":1067},9,[1069,1074,1078,1082,1086,1090,1094],{"type":44,"tag":547,"props":1070,"children":1071},{"style":653},[1072],{"type":49,"value":1073},"pip",{"type":44,"tag":547,"props":1075,"children":1076},{"style":597},[1077],{"type":49,"value":671},{"type":44,"tag":547,"props":1079,"children":1080},{"style":597},[1081],{"type":49,"value":676},{"type":44,"tag":547,"props":1083,"children":1084},{"style":566},[1085],{"type":49,"value":681},{"type":44,"tag":547,"props":1087,"children":1088},{"style":560},[1089],{"type":49,"value":686},{"type":44,"tag":547,"props":1091,"children":1092},{"style":597},[1093],{"type":49,"value":691},{"type":44,"tag":547,"props":1095,"children":1096},{"style":566},[1097],{"type":49,"value":919},{"type":44,"tag":488,"props":1099,"children":1100},{},[1101],{"type":49,"value":1102},"Download the MR-brain weights:",{"type":44,"tag":536,"props":1104,"children":1106},{"className":538,"code":1105,"language":540,"meta":541,"style":541},"cd \"$NV_GENERATE_ROOT\"\npython -m scripts.download_model_data --version rflow-mr-brain --root_dir .\u002F --model_only\n",[1107],{"type":44,"tag":83,"props":1108,"children":1109},{"__ignoreMap":541},[1110,1131],{"type":44,"tag":547,"props":1111,"children":1112},{"class":549,"line":550},[1113,1119,1123,1127],{"type":44,"tag":547,"props":1114,"children":1116},{"style":1115},"--shiki-light:#6182B8;--shiki-default:#82AAFF;--shiki-dark:#82AAFF",[1117],{"type":49,"value":1118},"cd",{"type":44,"tag":547,"props":1120,"children":1121},{"style":566},[1122],{"type":49,"value":681},{"type":44,"tag":547,"props":1124,"children":1125},{"style":560},[1126],{"type":49,"value":686},{"type":44,"tag":547,"props":1128,"children":1129},{"style":566},[1130],{"type":49,"value":919},{"type":44,"tag":547,"props":1132,"children":1133},{"class":549,"line":649},[1134,1138,1142,1147,1152,1157,1162,1167],{"type":44,"tag":547,"props":1135,"children":1136},{"style":653},[1137],{"type":49,"value":656},{"type":44,"tag":547,"props":1139,"children":1140},{"style":597},[1141],{"type":49,"value":661},{"type":44,"tag":547,"props":1143,"children":1144},{"style":597},[1145],{"type":49,"value":1146}," scripts.download_model_data",{"type":44,"tag":547,"props":1148,"children":1149},{"style":597},[1150],{"type":49,"value":1151}," --version",{"type":44,"tag":547,"props":1153,"children":1154},{"style":597},[1155],{"type":49,"value":1156}," rflow-mr-brain",{"type":44,"tag":547,"props":1158,"children":1159},{"style":597},[1160],{"type":49,"value":1161}," --root_dir",{"type":44,"tag":547,"props":1163,"children":1164},{"style":597},[1165],{"type":49,"value":1166}," .\u002F",{"type":44,"tag":547,"props":1168,"children":1169},{"style":597},[1170],{"type":49,"value":1171}," --model_only\n",{"type":44,"tag":488,"props":1173,"children":1174},{},[1175],{"type":49,"value":1176},"Runtime needs an NVIDIA GPU with at least 16 GB VRAM. There is no CPU\nfallback in the upstream path.",{"type":44,"tag":488,"props":1178,"children":1179},{},[1180,1182,1188,1190,1195],{"type":49,"value":1181},"The wrapper also searches ",{"type":44,"tag":83,"props":1183,"children":1185},{"className":1184},[],[1186],{"type":49,"value":1187},".workbench_data\u002Fupstreams\u002FNV-Generate-CTMR",{"type":49,"value":1189}," if\n",{"type":44,"tag":83,"props":1191,"children":1193},{"className":1192},[],[1194],{"type":49,"value":579},{"type":49,"value":1196}," is unset or does not have the required upstream layout.",{"type":44,"tag":488,"props":1198,"children":1199},{},[1200,1202,1208,1210,1216,1218,1224],{"type":49,"value":1201},"For agent-generated user run commands, use the command in Usage. Do not prepend\nclone or model-download setup steps when the repo-local\nupstream cache already exists. In a fresh Python environment, still include\n",{"type":44,"tag":83,"props":1203,"children":1205},{"className":1204},[],[1206],{"type":49,"value":1207},"pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\"",{"type":49,"value":1209}," before the wrapper unless\nthe active environment has already proven those imports are available; cached\nweights do not imply cached Python packages. If setup requires ",{"type":44,"tag":83,"props":1211,"children":1213},{"className":1212},[],[1214],{"type":49,"value":1215},"cd \"$NV_GENERATE_ROOT\"",{"type":49,"value":1217},", return to the Medical AI Skills repo before invoking\n",{"type":44,"tag":83,"props":1219,"children":1221},{"className":1220},[],[1222],{"type":49,"value":1223},"skills\u002Fnv-generate-mr-brain\u002Fscripts\u002Frun_mr_brain.py",{"type":49,"value":106},{"type":44,"tag":52,"props":1226,"children":1228},{"id":1227},"usage",[1229],{"type":49,"value":1230},"Usage",{"type":44,"tag":536,"props":1232,"children":1234},{"className":538,"code":1233,"language":540,"meta":541,"style":541},"export NV_GENERATE_ROOT=\"${NV_GENERATE_ROOT:-$HOME\u002F.cache\u002Fnvidia-skills\u002Fupstreams\u002FNV-Generate-CTMR-61c4ec7}\" && \\\npython -m pip install -r \"$NV_GENERATE_ROOT\u002Frequirements.txt\" && \\\npython skills\u002Fnv-generate-mr-brain\u002Fscripts\u002Frun_mr_brain.py \\\n  PATH_TO_MR_BRAIN_CONFIG.json \\\n  --output-dir runs\u002Fnv_generate_mr_brain_demo \\\n  --modality mri_t1 \\\n  --random-seed 1234\n",[1235],{"type":44,"tag":83,"props":1236,"children":1237},{"__ignoreMap":541},[1238,1317,1364,1379,1391,1408,1424],{"type":44,"tag":547,"props":1239,"children":1240},{"class":549,"line":550},[1241,1245,1249,1253,1257,1261,1265,1269,1273,1277,1281,1285,1289,1293,1297,1301,1305,1309,1313],{"type":44,"tag":547,"props":1242,"children":1243},{"style":554},[1244],{"type":49,"value":557},{"type":44,"tag":547,"props":1246,"children":1247},{"style":560},[1248],{"type":49,"value":563},{"type":44,"tag":547,"props":1250,"children":1251},{"style":566},[1252],{"type":49,"value":569},{"type":44,"tag":547,"props":1254,"children":1255},{"style":566},[1256],{"type":49,"value":574},{"type":44,"tag":547,"props":1258,"children":1259},{"style":560},[1260],{"type":49,"value":579},{"type":44,"tag":547,"props":1262,"children":1263},{"style":566},[1264],{"type":49,"value":584},{"type":44,"tag":547,"props":1266,"children":1267},{"style":560},[1268],{"type":49,"value":589},{"type":44,"tag":547,"props":1270,"children":1271},{"style":566},[1272],{"type":49,"value":594},{"type":44,"tag":547,"props":1274,"children":1275},{"style":597},[1276],{"type":49,"value":106},{"type":44,"tag":547,"props":1278,"children":1279},{"style":560},[1280],{"type":49,"value":604},{"type":44,"tag":547,"props":1282,"children":1283},{"style":566},[1284],{"type":49,"value":594},{"type":44,"tag":547,"props":1286,"children":1287},{"style":560},[1288],{"type":49,"value":613},{"type":44,"tag":547,"props":1290,"children":1291},{"style":566},[1292],{"type":49,"value":594},{"type":44,"tag":547,"props":1294,"children":1295},{"style":560},[1296],{"type":49,"value":622},{"type":44,"tag":547,"props":1298,"children":1299},{"style":566},[1300],{"type":49,"value":594},{"type":44,"tag":547,"props":1302,"children":1303},{"style":560},[1304],{"type":49,"value":631},{"type":44,"tag":547,"props":1306,"children":1307},{"style":566},[1308],{"type":49,"value":636},{"type":44,"tag":547,"props":1310,"children":1311},{"style":566},[1312],{"type":49,"value":641},{"type":44,"tag":547,"props":1314,"children":1315},{"style":560},[1316],{"type":49,"value":646},{"type":44,"tag":547,"props":1318,"children":1319},{"class":549,"line":649},[1320,1324,1328,1332,1336,1340,1344,1348,1352,1356,1360],{"type":44,"tag":547,"props":1321,"children":1322},{"style":653},[1323],{"type":49,"value":656},{"type":44,"tag":547,"props":1325,"children":1326},{"style":597},[1327],{"type":49,"value":661},{"type":44,"tag":547,"props":1329,"children":1330},{"style":597},[1331],{"type":49,"value":666},{"type":44,"tag":547,"props":1333,"children":1334},{"style":597},[1335],{"type":49,"value":671},{"type":44,"tag":547,"props":1337,"children":1338},{"style":597},[1339],{"type":49,"value":676},{"type":44,"tag":547,"props":1341,"children":1342},{"style":566},[1343],{"type":49,"value":681},{"type":44,"tag":547,"props":1345,"children":1346},{"style":560},[1347],{"type":49,"value":686},{"type":44,"tag":547,"props":1349,"children":1350},{"style":597},[1351],{"type":49,"value":691},{"type":44,"tag":547,"props":1353,"children":1354},{"style":566},[1355],{"type":49,"value":696},{"type":44,"tag":547,"props":1357,"children":1358},{"style":566},[1359],{"type":49,"value":641},{"type":44,"tag":547,"props":1361,"children":1362},{"style":560},[1363],{"type":49,"value":646},{"type":44,"tag":547,"props":1365,"children":1366},{"class":549,"line":707},[1367,1371,1375],{"type":44,"tag":547,"props":1368,"children":1369},{"style":653},[1370],{"type":49,"value":656},{"type":44,"tag":547,"props":1372,"children":1373},{"style":597},[1374],{"type":49,"value":717},{"type":44,"tag":547,"props":1376,"children":1377},{"style":560},[1378],{"type":49,"value":646},{"type":44,"tag":547,"props":1380,"children":1381},{"class":549,"line":922},[1382,1387],{"type":44,"tag":547,"props":1383,"children":1384},{"style":597},[1385],{"type":49,"value":1386},"  PATH_TO_MR_BRAIN_CONFIG.json",{"type":44,"tag":547,"props":1388,"children":1389},{"style":560},[1390],{"type":49,"value":646},{"type":44,"tag":547,"props":1392,"children":1393},{"class":549,"line":970},[1394,1399,1404],{"type":44,"tag":547,"props":1395,"children":1396},{"style":597},[1397],{"type":49,"value":1398},"  --output-dir",{"type":44,"tag":547,"props":1400,"children":1401},{"style":597},[1402],{"type":49,"value":1403}," runs\u002Fnv_generate_mr_brain_demo",{"type":44,"tag":547,"props":1405,"children":1406},{"style":560},[1407],{"type":49,"value":646},{"type":44,"tag":547,"props":1409,"children":1410},{"class":549,"line":1001},[1411,1416,1420],{"type":44,"tag":547,"props":1412,"children":1413},{"style":597},[1414],{"type":49,"value":1415},"  --modality",{"type":44,"tag":547,"props":1417,"children":1418},{"style":597},[1419],{"type":49,"value":742},{"type":44,"tag":547,"props":1421,"children":1422},{"style":560},[1423],{"type":49,"value":646},{"type":44,"tag":547,"props":1425,"children":1426},{"class":549,"line":1049},[1427,1432],{"type":44,"tag":547,"props":1428,"children":1429},{"style":597},[1430],{"type":49,"value":1431},"  --random-seed",{"type":44,"tag":547,"props":1433,"children":1434},{"style":750},[1435],{"type":49,"value":753},{"type":44,"tag":488,"props":1437,"children":1438},{},[1439,1441,1446,1448,1454,1456,1461],{"type":49,"value":1440},"Replace ",{"type":44,"tag":83,"props":1442,"children":1444},{"className":1443},[],[1445],{"type":49,"value":794},{"type":49,"value":1447}," with the user's actual request\u002Fconfig\npath. Do not copy the fixture path from this document unless the user\nexplicitly asked to run that fixture. If the user says \"the request is at\n",{"type":44,"tag":83,"props":1449,"children":1451},{"className":1450},[],[1452],{"type":49,"value":1453},"runs\u002F...\u002Fdefault_mri_t1.json",{"type":49,"value":1455},"\", that exact path is the first positional\nargument to ",{"type":44,"tag":83,"props":1457,"children":1459},{"className":1458},[],[1460],{"type":49,"value":139},{"type":49,"value":106},{"type":44,"tag":488,"props":1463,"children":1464},{},[1465,1467,1473,1474,1480,1481,1487,1489,1495,1496,1502,1503,1509,1510,1516,1517,1523,1525,1531,1533,1539,1541,1547],{"type":49,"value":1466},"Supported MR-brain modality names are ",{"type":44,"tag":83,"props":1468,"children":1470},{"className":1469},[],[1471],{"type":49,"value":1472},"mri",{"type":49,"value":203},{"type":44,"tag":83,"props":1475,"children":1477},{"className":1476},[],[1478],{"type":49,"value":1479},"mri_t1",{"type":49,"value":203},{"type":44,"tag":83,"props":1482,"children":1484},{"className":1483},[],[1485],{"type":49,"value":1486},"mri_t2",{"type":49,"value":1488},",\n",{"type":44,"tag":83,"props":1490,"children":1492},{"className":1491},[],[1493],{"type":49,"value":1494},"mri_flair",{"type":49,"value":203},{"type":44,"tag":83,"props":1497,"children":1499},{"className":1498},[],[1500],{"type":49,"value":1501},"mri_swi",{"type":49,"value":203},{"type":44,"tag":83,"props":1504,"children":1506},{"className":1505},[],[1507],{"type":49,"value":1508},"mri_t1_skull_stripped",{"type":49,"value":1488},{"type":44,"tag":83,"props":1511,"children":1513},{"className":1512},[],[1514],{"type":49,"value":1515},"mri_t2_skull_stripped",{"type":49,"value":203},{"type":44,"tag":83,"props":1518,"children":1520},{"className":1519},[],[1521],{"type":49,"value":1522},"mri_flair_skull_stripped",{"type":49,"value":1524},", and\n",{"type":44,"tag":83,"props":1526,"children":1528},{"className":1527},[],[1529],{"type":49,"value":1530},"mri_swi_skull_stripped",{"type":49,"value":1532},". These map to the upstream\n",{"type":44,"tag":83,"props":1534,"children":1536},{"className":1535},[],[1537],{"type":49,"value":1538},"configs\u002Fmodality_mapping.json",{"type":49,"value":1540}," IDs documented in the README.\nFor FOV and setup details, see ",{"type":44,"tag":83,"props":1542,"children":1544},{"className":1543},[],[1545],{"type":49,"value":1546},"references\u002Ffov-and-downloads.md",{"type":49,"value":106},{"type":44,"tag":488,"props":1549,"children":1550},{},[1551,1553,1559,1561,1567,1569,1575,1576,1582,1583,1589,1590,1596,1597,1603],{"type":49,"value":1552},"The fixture argument is a small JSON override for\n",{"type":44,"tag":83,"props":1554,"children":1556},{"className":1555},[],[1557],{"type":49,"value":1558},"configs\u002Fconfig_maisi_diff_model_rflow-mr-brain.json",{"type":49,"value":1560},". Pass ",{"type":44,"tag":83,"props":1562,"children":1564},{"className":1563},[],[1565],{"type":49,"value":1566},"default",{"type":49,"value":1568}," to use\nthe upstream defaults plus the CLI modality and random seed. Common override\nkeys are ",{"type":44,"tag":83,"props":1570,"children":1572},{"className":1571},[],[1573],{"type":49,"value":1574},"dim",{"type":49,"value":203},{"type":44,"tag":83,"props":1577,"children":1579},{"className":1578},[],[1580],{"type":49,"value":1581},"spacing",{"type":49,"value":203},{"type":44,"tag":83,"props":1584,"children":1586},{"className":1585},[],[1587],{"type":49,"value":1588},"num_inference_steps",{"type":49,"value":203},{"type":44,"tag":83,"props":1591,"children":1593},{"className":1592},[],[1594],{"type":49,"value":1595},"cfg_guidance_scale",{"type":49,"value":1524},{"type":44,"tag":83,"props":1598,"children":1600},{"className":1599},[],[1601],{"type":49,"value":1602},"modality",{"type":49,"value":106},{"type":44,"tag":488,"props":1605,"children":1606},{},[1607],{"type":49,"value":1608},"Each run records the staged config, model inventory, upstream command, output\ngeometry, spacing, affine, intensity range, and non-constant \u002F finite-data\nchecks. Output volumes are synthetic and are not safe as production training\ndata without independent review.",{"type":44,"tag":488,"props":1610,"children":1611},{},[1612],{"type":49,"value":1613},"Not for clinical interpretation, production deployment, autonomous diagnosis,\nor regulatory submission.",{"type":44,"tag":1615,"props":1616,"children":1617},"style",{},[1618],{"type":49,"value":1619},"html .light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html.light .shiki span {color: var(--shiki-light);background: var(--shiki-light-bg);font-style: var(--shiki-light-font-style);font-weight: var(--shiki-light-font-weight);text-decoration: var(--shiki-light-text-decoration);}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"items":1621,"total":1723},[1622,1639,1653,1667,1679,1694,1709],{"slug":1623,"name":1623,"fn":1624,"description":1625,"org":1626,"tags":1627,"stars":20,"repoUrl":21,"updatedAt":1638},"accelerated-computing-cudf","accelerate data processing with cuDF","Official NVIDIA-authored guidance for NVIDIA cuDF GPU DataFrames, pandas acceleration, dask-cuDF, ETL, joins, groupby, CSV\u002FParquet I\u002FO, nullable semantics, and multi-GPU DataFrame workloads.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1628,1631,1634,1635],{"name":1629,"slug":1630,"type":15},"Data Analysis","data-analysis",{"name":1632,"slug":1633,"type":15},"Data Engineering","data-engineering",{"name":9,"slug":8,"type":15},{"name":1636,"slug":1637,"type":15},"Performance","performance","2026-07-14T05:28:43.176466",{"slug":1640,"name":1640,"fn":1641,"description":1642,"org":1643,"tags":1644,"stars":20,"repoUrl":21,"updatedAt":1652},"aiq-deploy","deploy and manage NVIDIA AI-Q infrastructure","Use when asked to install, deploy, run, validate, troubleshoot, or stop NVIDIA AI-Q Blueprint infrastructure.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1645,1648,1651],{"name":1646,"slug":1647,"type":15},"Deployment","deployment",{"name":1649,"slug":1650,"type":15},"Infrastructure","infrastructure",{"name":9,"slug":8,"type":15},"2026-07-14T05:29:06.667109",{"slug":1654,"name":1654,"fn":1655,"description":1656,"org":1657,"tags":1658,"stars":20,"repoUrl":21,"updatedAt":1666},"aiq-research","conduct deep research with AI-Q","Use when asked to run deep research or AI-Q research through a reachable NVIDIA AI-Q Blueprint backend.\n",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1659,1662,1663],{"name":1660,"slug":1661,"type":15},"Agents","agents",{"name":9,"slug":8,"type":15},{"name":1664,"slug":1665,"type":15},"Research","research","2026-07-14T05:28:06.816956",{"slug":1668,"name":1668,"fn":1669,"description":1670,"org":1671,"tags":1672,"stars":20,"repoUrl":21,"updatedAt":1678},"amc-run-sample-calibration","run AMC sample dataset calibration","Run end-to-end calibration on the shipped sample dataset (sdg_08_2_sample_data_010926.zip) against a running AMC microservice. Use when user says 'test sample dataset', 'run sample calibration', 'verify AMC install', or 'launch and test'.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1673,1674,1675],{"name":1629,"slug":1630,"type":15},{"name":9,"slug":8,"type":15},{"name":1676,"slug":1677,"type":15},"Testing","testing","2026-07-17T05:29:03.913266",{"slug":1680,"name":1680,"fn":1681,"description":1682,"org":1683,"tags":1684,"stars":20,"repoUrl":21,"updatedAt":1693},"amc-run-video-calibration","calibrate video datasets with AutoMagicCalib","Calibrate a new dataset from pre-recorded video files via the AutoMagicCalib REST API. Use when user has local MP4s and says 'calibrate my videos', 'run AMC on these videos', or similar. For RTSP\u002Flive streams, use amc-run-rtsp-calibration instead.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1685,1688,1689,1690],{"name":1686,"slug":1687,"type":15},"Automation","automation",{"name":18,"slug":19,"type":15},{"name":9,"slug":8,"type":15},{"name":1691,"slug":1692,"type":15},"Video","video","2026-07-17T05:28:53.905004",{"slug":1695,"name":1695,"fn":1696,"description":1697,"org":1698,"tags":1699,"stars":20,"repoUrl":21,"updatedAt":1708},"amc-setup-calibration-stack","deploy AutoMagicCalib microservice with Docker","Launch AutoMagicCalib microservice and web UI from NGC release images via Docker Compose. Use when user says 'deploy auto calibration', 'launch auto calibration', 'launch AMC', 'start MS+UI', or 'set up auto-magic-calib'. Requires NGC API key.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1700,1701,1704,1705],{"name":1646,"slug":1647,"type":15},{"name":1702,"slug":1703,"type":15},"Docker","docker",{"name":9,"slug":8,"type":15},{"name":1706,"slug":1707,"type":15},"Operations","operations","2026-07-17T05:28:56.913999",{"slug":1710,"name":1710,"fn":1711,"description":1712,"org":1713,"tags":1714,"stars":20,"repoUrl":21,"updatedAt":1722},"cudaq-guide","develop quantum applications with CUDA-Q","CUDA-Q onboarding guide for installation, test programs, GPU simulation, QPU hardware, and quantum applications.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1715,1716,1719],{"name":9,"slug":8,"type":15},{"name":1717,"slug":1718,"type":15},"Quantum Computing","quantum-computing",{"name":1720,"slug":1721,"type":15},"Simulation","simulation","2026-07-14T05:26:58.898253",305,{"items":1725,"total":1875},[1726,1744,1759,1770,1782,1796,1809,1823,1834,1843,1857,1866],{"slug":1727,"name":1727,"fn":1728,"description":1729,"org":1730,"tags":1731,"stars":1741,"repoUrl":1742,"updatedAt":1743},"nemoclaw-user-guide","retrieve NemoClaw documentation and configuration","Guides human users' AI agents to the NemoClaw docs MCP server and canonical Fern documentation in Markdown form. Use when users ask how to install, configure, operate, troubleshoot, secure, or learn NemoClaw with an AI coding assistant. Trigger keywords - nemoclaw docs, use nemoclaw with ai agent, nemoclaw mcp docs, nemoclaw install help, nemoclaw quickstart, nemoclaw markdown docs, llms.txt, agent skills.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1732,1735,1738],{"name":1733,"slug":1734,"type":15},"Documentation","documentation",{"name":1736,"slug":1737,"type":15},"MCP","mcp",{"name":1739,"slug":1740,"type":15},"Search","search",21777,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FNemoClaw","2026-07-20T06:00:01.461044",{"slug":1745,"name":1745,"fn":1746,"description":1747,"org":1748,"tags":1749,"stars":1756,"repoUrl":1757,"updatedAt":1758},"mcore-build-and-dependency","manage Megatron-LM development environments","Container-based dev environment setup and dependency management for Megatron-LM. Covers acquiring and launching the CI container, uv package management, and updating uv.lock.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1750,1753,1754],{"name":1751,"slug":1752,"type":15},"Containers","containers",{"name":1646,"slug":1647,"type":15},{"name":1755,"slug":656,"type":15},"Python",17049,"https:\u002F\u002Fgithub.com\u002FNVIDIA\u002FMegatron-LM","2026-07-27T06:06:11.249662",{"slug":1760,"name":1760,"fn":1761,"description":1762,"org":1763,"tags":1764,"stars":1756,"repoUrl":1757,"updatedAt":1769},"mcore-bump-base-image","update NVIDIA PyTorch base images","Bump the NVIDIA PyTorch base image (`nvcr.io\u002Fnvidia\u002Fpytorch:YY.MM-py3`) used by Megatron-LM CI. Covers the two pin sites (GitHub CI in `docker\u002F.ngc_version.dev` and GitLab CI in `.gitlab\u002Fstages\u002F01.build.yml`), the post-bump CI loop (re-run functional tests, refresh golden values, mark broken tests), and the gotchas that bit PRs",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1765,1768],{"name":1766,"slug":1767,"type":15},"CI\u002FCD","ci-cd",{"name":1646,"slug":1647,"type":15},"2026-07-14T05:25:59.97109",{"slug":1771,"name":1771,"fn":1772,"description":1773,"org":1774,"tags":1775,"stars":1756,"repoUrl":1757,"updatedAt":1781},"mcore-cicd","manage CI\u002FCD pipelines for Megatron-LM","CI\u002FCD reference for Megatron-LM. Covers CI pipeline structure, PR scope labels, triggering internal GitLab CI (which force-pushes the current branch to a pull-request\u002FBRANCH ref — always dry-run and verify the destination first; never run against shared or protected branches), and CI failure investigation.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1776,1777,1778],{"name":1766,"slug":1767,"type":15},{"name":1646,"slug":1647,"type":15},{"name":1779,"slug":1780,"type":15},"GitHub","github","2026-07-27T06:06:12.278222",{"slug":1783,"name":1783,"fn":1784,"description":1785,"org":1786,"tags":1787,"stars":1756,"repoUrl":1757,"updatedAt":1795},"mcore-create-issue","investigate CI failures and create issues","Investigate a failing GitHub Actions run or job and create a GitHub issue for the failure.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1788,1791,1792],{"name":1789,"slug":1790,"type":15},"Debugging","debugging",{"name":1779,"slug":1780,"type":15},{"name":1793,"slug":1794,"type":15},"Triage","triage","2026-07-14T05:25:57.442089",{"slug":1797,"name":1797,"fn":1798,"description":1799,"org":1800,"tags":1801,"stars":1756,"repoUrl":1757,"updatedAt":1808},"mcore-linting-and-formatting","lint and format Megatron-LM code","Linting and formatting for Megatron-LM. Covers running autoformat.sh, tools (ruff, black, isort, pylint, mypy), and code style rules.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1802,1805],{"name":1803,"slug":1804,"type":15},"Best Practices","best-practices",{"name":1806,"slug":1807,"type":15},"Code Analysis","code-analysis","2026-07-14T05:25:56.18433",{"slug":1810,"name":1810,"fn":1811,"description":1812,"org":1813,"tags":1814,"stars":1756,"repoUrl":1757,"updatedAt":1822},"mcore-migrate-gpt-to-hybrid","migrate Megatron-LM models to HybridModel","Migration guide for moving Megatron Core GPTModel checkpoints, model providers, training commands, and layer mappings to HybridModel.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1815,1818,1821],{"name":1816,"slug":1817,"type":15},"Machine Learning","machine-learning",{"name":1819,"slug":1820,"type":15},"Migration","migration",{"name":9,"slug":8,"type":15},"2026-07-17T06:07:11.777011",{"slug":1824,"name":1824,"fn":1825,"description":1826,"org":1827,"tags":1828,"stars":1756,"repoUrl":1757,"updatedAt":1833},"mcore-onboard-gb200-1node-tests","onboard functional tests for GB200","Onboard 1-node GitHub MR functional tests for GB200 from existing mr-scoped 2-node tests.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1829,1832],{"name":1830,"slug":1831,"type":15},"QA","qa",{"name":1676,"slug":1677,"type":15},"2026-07-14T05:25:53.673039",{"slug":1835,"name":1835,"fn":1836,"description":1837,"org":1838,"tags":1839,"stars":1756,"repoUrl":1757,"updatedAt":1842},"mcore-run-on-slurm","launch distributed training jobs on SLURM","How to launch distributed Megatron-LM training jobs on a SLURM cluster. Covers a minimal sbatch skeleton, environment-variable setup for torch.distributed.run, CUDA_DEVICE_MAX_CONNECTIONS rules across hardware and parallelism modes, container conventions, monitoring, and per-rank failure diagnosis.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1840,1841],{"name":1646,"slug":1647,"type":15},{"name":1649,"slug":1650,"type":15},"2026-07-14T05:25:49.362534",{"slug":1844,"name":1844,"fn":1845,"description":1846,"org":1847,"tags":1848,"stars":1756,"repoUrl":1757,"updatedAt":1856},"mcore-split-pr","split pull requests to reduce review load","Split a PR into multiple PRs to reduce the number of required CODEOWNERS reviewer groups.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1849,1852,1853],{"name":1850,"slug":1851,"type":15},"Code Review","code-review",{"name":1779,"slug":1780,"type":15},{"name":1854,"slug":1855,"type":15},"Pull Requests","pull-requests","2026-07-14T05:26:01.226578",{"slug":1858,"name":1858,"fn":1859,"description":1860,"org":1861,"tags":1862,"stars":1756,"repoUrl":1757,"updatedAt":1865},"mcore-testing","run and manage Megatron-LM tests","Test system for Megatron-LM. Covers test layout, recipe YAML structure, adding and running unit and functional tests, golden values, marker filters, and CI parity.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1863,1864],{"name":1830,"slug":1831,"type":15},{"name":1676,"slug":1677,"type":15},"2026-07-14T05:25:54.928983",{"slug":1867,"name":1867,"fn":1868,"description":1869,"org":1870,"tags":1871,"stars":1756,"repoUrl":1757,"updatedAt":1874},"nightly-sync","manage nightly main-to-dev sync workflows","Domain knowledge for the nightly main-to-dev sync workflow. Covers merge strategy, CI architecture, failure investigation, and known issues.",{"slug":8,"name":9,"logoUrl":10,"githubOrg":9},[1872,1873],{"name":1686,"slug":1687,"type":15},{"name":1766,"slug":1767,"type":15},"2026-07-30T05:29:03.275638",496]