Dev & Engineering ✓ NVIDIA · Official roboticspolicy-finetuninglerobotgr00topenpi-pi0gpu-trainingisaac-for-healthcare

I4H Policy Fine-Tuning

Fine-tune GR00T or openpi PI0 policies on recorded LeRobot demonstrations.

FollowSkills review · FSRS-2.0
Not recommended
48/ 100 5-point scale 2.4 / 5
Trust12 / 25 · 2.4/5

The skill scopes work to a local dataset and GPU and does not request credentials or explicitly export data. However, it clones a remote repository, creates directories and a symlink, and starts training without explicit confirmation; permission boundaries, rollback, and data-flow disclosure are incomplete, so points are deducted.

Reliability8 / 20 · 2.0/5

The workflow includes environment mappings, preflight checks, verification criteria, and common troubleshooting. However, it depends on an external repository, uv, GPUs, and assumed tmux variable persistence, with no executable tests for this skill's key paths. Static calibration therefore limits the score below 10.

Adaptability10 / 15 · 3.3/5

The audience, trigger wording, supported environments, and non-training boundaries are reasonably clear. Coverage is limited to four environments, input boundaries are incomplete, and the default GitHub clone and overseas toolchain may be unreachable from mainland China; Chinese-language support is also unspecified, so points are deducted.

Convention9 / 15 · 3.0/5

The material provides a name, version, Apache-2.0 license, author, layered procedure, limitations, troubleshooting, and final reporting requirements. It lacks the recommended Instructions and Examples sections, has limited changelog, maintenance, update-path, and parameter-stability information, and repository license metadata is NOASSERTION, so points are deducted.

Effectiveness6 / 15 · 2.0/5

The skill can directly guide dataset preflight, training, checkpoint inspection, and reporting of key results, and the benchmark reports some benefit on two positive tasks. However, the benchmark has only two single-attempt tasks and lacks representative usable outputs and failure-case evidence; the static score is conservatively limited.

Verifiability3 / 10 · 1.5/5

Versioned documentation, an evaluation summary, and internal evaluation tasks provide limited auditability. There are no raw logs, independent reproduction artifacts, negative-trigger cases, or committed tests covering this skill's key paths, so evidence remains thin.

Evidence confidence:Low Reviewed Jul 29, 2026 Reviewed revision ce70ca7f1966
The upstream repository has new commits since this review. The score still applies to the reviewed revision shown and may not cover the latest changes.
Before you use it
  • The default flow may clone remote code and launch GPU training without explicit user confirmation; confirm repository provenance, network access, disk space, GPU resources, and output paths first.
  • Fine-tuning healthcare-robotics policies can involve sensitive data and real-world safety risks; the document lacks guidance on de-identification, isolation, model safety evaluation, and recovery after failed training.
  • The benchmark covers only two positive tasks with one attempt each and does not establish boundary-trigger accuracy, abnormal-input handling, or cross-environment stability.
See the full review method →

What does this skill do, and when should you use it?

This skill guides Isaac for Healthcare workflows for fine-tuning a policy on an existing LeRobot dataset of demonstrations. It supports GR00T and openpi PI0, selecting a training stack and CLI for the chosen environment. The workflow resolves the base code, checks the dataset and GPUs, runs training, and validates the resulting checkpoint and logs. It is for training rather than checkpoint evaluation, and inference-only environments cannot be fine-tuned.

Resolves or clones the i4h-workflows base repository; reads meta/info.json from a LeRobot dataset; checks visible GPU count and available environments; maps each supported environment to the gr00t_n15, gr00t_n16, or openpi_pi0 stack and training CLI; runs training through uv; writes checkpoints and logs under workflows/agentic/runs; and checks for safetensors files, experiment_cfg, processor, train_loss entries, and the final training summary.

  1. A robotics researcher needs to fine-tune a GR00T N1.5 policy on scissor_pick_and_place demonstrations.
  2. An embodied-AI developer needs to train a GR00T N1.6 policy for locomanip_tray_pick_and_place or locomanip_push_cart.
  3. A healthcare robotics team needs to train an openpi PI0 policy for ultrasound_liver_scan.
  4. A user with a small dataset wants to freeze the vision backbone to reduce memory use and training time.

What are this skill's strengths and limitations?

Pros
  • Provides explicit environment-to-stack-to-CLI mappings.
  • Includes preflight and post-run checks for datasets, GPUs, training support, checkpoints, and logs.
  • Exposes practical controls for batch size, learning rate, training steps, GPU count, and vision-backbone freezing.
  • Documents common setup, dataset, environment, and CLI-flag failures.
Limitations
  • Requires GPU hardware, with --num-gpus limited by visible GPUs.
  • Requires an existing LeRobot directory containing meta/info.json.
  • Environments with null policy.train_module, including assemble_trocar, cannot be fine-tuned.
  • The source provides no evidence about training quality, hardware coverage, or a standalone test suite.

How do you install this skill?

Install the NVIDIA skills collection with: npx skills add nvidia/skills. Install only this skill with: npx skills add nvidia/skills --skill i4h-workflow-finetune --yes. The README does not document a separate installation directory or additional skill-specific installation procedure.

How do you use this skill?

Ask a loaded agent to fine-tune a policy on recorded LeRobot demonstrations. Prepare the i4h-workflows setup, a LeRobot dataset containing meta/info.json, and at least one visible NVIDIA GPU. Set DATASET_PATH, then run the documented preflight and training commands in order. Supported environments are scissor_pick_and_place, locomanip_tray_pick_and_place, locomanip_push_cart, and ultrasound_liver_scan.

How does this skill compare with similar options?

The skill explicitly distinguishes fine-tuning and training from checkpoint evaluation; evaluation should use i4h-workflow-validate.

FAQ

Can assemble_trocar be trained with this skill?
No. It is marked inference-only because its policy.train_module is null.
How many GPUs are required?
At least one visible GPU is required, and --num-gpus cannot exceed the number reported by nvidia-smi.
What dataset format is required?
The input must be an existing LeRobot dataset directory containing meta/info.json.
What should I check when training fails?
Verify that i4h-workflows and the relevant stack virtual environment are set up, DATASET_PATH is valid, the environment supports training, and flags use kebab case such as --max-steps.

More skills from this repository

All from NVIDIA/skills

Data & Analysis ✓ NVIDIA · Official

I4H Dataset Converter

Convert Isaac for Healthcare HDF5 recordings into training-ready LeRobot datasets.

Data & Analysis ✓ NVIDIA · Official

LeRobot Dataset Visualizer

Inspect converted LeRobot datasets in a local browser.

Dev & Engineering ✓ NVIDIA · Official

I4H Workflow Environment Creator

Scaffold a runnable Isaac for Healthcare environment by forking a proven existing environment.

Dev & Engineering ✓ NVIDIA · Official

Isaac for Healthcare Dataset Replay

Replay HDF5 episodes in Isaac Sim to verify recorded robot behavior visually.

Data & Analysis ✓ NVIDIA · Official

I4H Teleoperation Dataset Recorder

Record human demonstrations from keyboard, SO-ARM leader, or VR teleoperation into HDF5.

Dev & Engineering ✓ NVIDIA · Official

i4h Agentic Workflow Guide

Navigate supported medical-robotics environments, policies, and workflow stages.

Dev & Engineering ✓ NVIDIA · Official

Isaac for Healthcare Workflow Validator

Run policy or state-machine rollouts and record verification episodes to HDF5.

Dev & Engineering ✓ NVIDIA · Official

Catheter Navigation Setup Verification

Checks host, GPU, and Python path readiness for catheter navigation.

Data & Analysis ✓ NVIDIA · Official

I4H Mimic Dataset Augmentation

Augment an existing HDF5 recording by cloning trajectories with action and state noise.

Dev & Engineering ✓ NVIDIA · Official

I4H Scene Editing Workflow

Edit existing Isaac for Healthcare scenes live through a controlled bridge session.

Dev & Engineering ✓ NVIDIA · Official

Catheter Navigation Fluoroscopy Viewport

Launch an interactive fluoroscopy viewport with real-time catheter physics and navigation controls.

Dev & Engineering ✓ NVIDIA · Official

Catheter Navigation E2E Smoke Test

Runs the Isaac for Healthcare v0.7 catheter workflow from setup verification through digital-twin DRR rendering and unit tests.

Automation & Ops ✓ NVIDIA · Official

Isaac for Healthcare E2E Workflow

Run the Isaac for Healthcare pipeline from recording through validation.

Data & Analysis ✓ NVIDIA · Official

i4H Dataset Annotation Workflow

Use a vision-language model to verify task success and filter episodes for fine-tuning.

Dev & Engineering ✓ NVIDIA · Official

TAO Platform Execution SDK

Submit, monitor, and manage NVIDIA TAO GPU training jobs across supported platforms.

Dev & Engineering ✓ NVIDIA · Official

NeMo-RL Auto Research

Turns NeMo-RL research goals into reproducible, iterative experiments.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge Training Guide

Run Megatron-LM and Megatron Bridge training and verify whether their loss curves match.

Dev & Engineering ✓ NVIDIA · Official

NeMo MBridge GPU Memory Tuning

Diagnose GPU OOMs in Megatron Bridge training and reduce fragmentation, activation, and PEFT memory usage.

Dev & Engineering ✓ NVIDIA · Official

Megatron Bridge Activation Recompute

Trade extra compute for lower GPU activation memory in Megatron Bridge training.

Automation & Ops ✓ NVIDIA · Official

Megatron-LM on SLURM

Launch, monitor, and troubleshoot multi-node Megatron-LM training on SLURM GPU clusters.

Related skills