Advanced workflows

Start with a successful PA → DPD experiment. These guides add a specific research or deployment workflow; they are not prerequisites for a first run.

Goal	Guide	What it covers
Fit classical MP/GMP baselines	Classical baselines	Least squares, ILA and fit stability.
Add a backbone	Adding a model	Register a model once for CLI, API and GUI.
Evaluate a known waveform	Waveform evaluation	Reference binding, data-aided metrics and their limitations.
Validate on a physical PA	Measured DPD	Capture provenance, alignment, with/without-DPD comparisons.
Compare adaptation strategies	Adaptation experiments	Zero update, few-shot and full retraining, with costs and failures.
Process consecutive chunks	Streaming	State, look-ahead, latency and chunk consistency.
Export a fixed-point model	Deployment export	Golden vectors and a verified C99 reference package.
Share benchmark evidence	Leaderboard submission	Packages, recomputation, review and versioned boards.

For comparable results, read the benchmark protocol and metric profiles. Current verification evidence is recorded in the support matrix, performance report and hardening report.