physical-ai-video-data-augmentation

Orchestrate NVIDIA OSmo VDA workflows with preflight checks and cache management.

Updated Apr 16, 2026

Search Tags:#workflow #orchestration #preflight #cache-management #osmo #vda #dataset-storage

Ready to run

npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill physical-ai-video-data-augmentation

💬 Prompt for ChatGPT / Claude Chat▼

Please help me install this Agent Skill.
Skill: physical-ai-video-data-augmentation
Source: https://github.com/sayalinvidia/sayali-skills-test/tree/main/skills/physical-ai-video-data-augmentation
Command: npx skills add https://github.com/sayalinvidia/sayali-skills-test --skill physical-ai-video-data-augmentation

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires pyyaml, omegaconf, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill coordinates the end-to-end orchestration of NVIDIA's Video Data Augmentation (VDA) workflows on OSMO, enabling safe preflight checks, dataset and model cache setup, and deterministic submit-time interpolation across flows.

Core Features & Use Cases

Orchestrates setup, augmentation, auto-labeling, and end-to-end pipelines with a single submission surface.
Manages prerequisites: credential wiring, cache prep, and pre-submit guard checks to reduce run-time errors.
Provides post-run artifact staging, monitoring, and evidence collection to assess augmentation impact.

Quick Start

Provide dataset, run_id, storage_url, and gpu_platform, then submit a VDA workflow using a single --set-string payload.

Frequently Asked Questions about physical-ai-video-data-augmentation

High-intent search queries and answers about installing and using this skill.

FAQPage Schema

How do I orchestrate video data augmentation workflows on NVIDIA OSMO?▼

Orchestrating VDA workflows on NVIDIA OSMO involves coordinating preflight checks, dataset cache setup, and submit-time interpolation using a single --set-string payload submission contract to ensure safe end-to-end runs.

What is submit-time interpolation in VDA pipeline orchestration?▼

Submit-time interpolation in VDA orchestration is the deterministic process of resolving configuration variables at run submission, enforcing a single --set-string payload contract to safely interpolate parameters across auto-labeling, augmentation, and E2E flows.

Can I use s3 or azure backends for dataset storage with OSMO VDA workflows?▼

Yes, OSMO VDA workflows support dataset storage across s3, azure, and gs backends, validating storage URLs and handling credentials to safely manage datasets stored in various cloud storage_url locations.

How do I run preflight checks before submitting an augmentation pipeline?▼

Preflight checks are executed during orchestration setup to validate dataset URLs, verify credential wiring, and prepare model caches, reducing run-time errors before you submit the VDA workflow payload.

What are the prerequisites for running end-to-end VDA workflows?▼

Prerequisites for running VDA workflows include providing the dataset, run_id, storage_url, and gpu_platform parameters, along with proper credential wiring and cache preparation managed through pre-submit guard checks.

Does OSMO VDA orchestration support auto-labeling flows?▼

Yes, OSMO VDA orchestration supports auto-labeling flows alongside augmentation and E2E pipelines, coordinating them through a single submission surface with strict guardrails and post-run artifact staging.