End-to-End Onboarding

Workflow Map

workflow

Capture to Screens

The tested RIA Hub path from upload or synthesis through curation, training, packaging, and a running Screens app.

1. Capture or synthesize Start from a local recording, uploaded RF sample, or synthetic signal problem.
2. Curate and label Select source files, slice useful intervals, and attach labels or qualifiers.
3. Inspect readiness Review balance, metadata, empty states, and recommendations before training.
4. Train model Choose a repository, runner, dataset, template, and mode to launch a workflow run.
5. Package app Connect model artifacts into an application pipeline and validate target runtime settings.
6. Run in Screens Start, stop, inspect logs, and view manifest panels in the Screens runtime.

capture

Capture or synthesize

Start from a local recording, uploaded RF sample, or synthetic signal problem.

Where this appears in RIA Hub

Generator, Library, and local capture workflows.

What you need next

A recording or generated dataset tracked in the Library.

Open guide

What You Will Build

End-to-end outputs

data Training-ready dataset A curated set of RF slices with labels, metadata, and inspection checks.train Model artifact A completed training run with metrics and an exportable model artifact.deploy Screens app A packaged app manifest that can start, stop, stream logs, and show runtime panels.

data

Training-ready dataset

This is the handoff from Library, Curator, and Inspector into Model Builder.

You will start with a recording or synthetic RF example, create a curated dataset, launch a training run, export a model artifact, package an application, and view it in Screens. The path mirrors the tested journey in the product: upload recording, curate, train, and export.

Prerequisites

Before you begin

Confirm access, storage, runner, and source-data assumptions before starting the walkthrough.

Open all Collapse all

Account and repository access access

Confirm you can access the RIA Hub instance, the target repository, and the branches used for data and model work.

Git LFS is available storage

Recordings, datasets, and generated artifacts can be large. Install Git LFS before cloning or pushing repositories that carry RF data.

Runner capacity exists runner

Training and packaging actions need an available runner with the required compute profile and permissions.

A signal problem is defined data

Start with a recording, a synthetic generator configuration, or a small modulation classification example.

terminal Local docs check

$ npm install
npm run dev

Open http://localhost:4321 and keep the docs beside the RIA Hub instance.

Outcome

Success means the Screens app is visible and has at least one panel showing inference or pipeline status.

Training run Ready artifact exported Screens app Running runtime controls available

Keep one artifact path visible

The model path selected after training is the same path the Application Composer and Screens packaging flow need later.

Next steps

Work through the tutorials in order to complete the full path:

1. Synthesis and Local Capture — create source RF data
2. Creating Recordings — generate SigMF recordings with the CLI
3. Curation and Labeling — build a labelled training dataset
4. Basic Training Materials — understand dataset requirements
5. Inspect a Dataset — verify dataset quality before training
6. Train a Model — launch a Model Builder run
7. Package an Application — connect artifacts into a deployable pipeline
8. Screens App — run the finished app