mirror of
https://github.com/ruvnet/RuView
synced 2026-06-21 12:13:19 +00:00
ruv
8ff7c2c35a
Add `plugins/ruview` — an end-to-end toolkit for working with RuView (WiFi-DensePose) from Claude Code, mirrored as Codex prompts. Marketplace: `plugins/.claude-plugin/marketplace.json` (one plugin, `ruview`). Skills (9): ruview-quickstart, ruview-hardware-setup, ruview-configure, ruview-applications, ruview-model-training, ruview-advanced-sensing, ruview-cli-api, ruview-mmwave, ruview-verify — shell-first (cargo / python / idf.py / docker / node), no claude-flow MCP dependency. Commands (7): /ruview-start, /ruview-flash, /ruview-provision, /ruview-app, /ruview-train, /ruview-advanced, /ruview-verify. Agents (3): ruview-onboarding-guide, ruview-config-engineer, ruview-training-engineer. Codex mirror: codex/AGENTS.md + codex/README.md + codex/prompts/*.md (full command parity, enforced by scripts/smoke.sh). Docs: docs/adrs/0001-ruview-plugin-contract.md (Proposed). Verification: scripts/smoke.sh (13 structural checks). Provisioning docs reflect the full `provision.py` flag set (TDM mesh, edge tiers, vitals, hop channels, Cognitum Seed, swarm intervals) and the issue #391 NVS-namespace-replace gotcha. Verified: `claude plugin validate` (plugin + marketplace), loads via `claude --plugin-dir`, smoke 13/13, and confirmed against an attached ESP32-S3 on COM8 running the RuView CSI firmware (live adaptive_ctrl + csi_collector serial output). Co-Authored-By: claude-flow <ruv@ruv.net>
19 lines
1.7 KiB
Markdown
19 lines
1.7 KiB
Markdown
---
|
|
description: Train a RuView model — camera-free WiFlow pose, camera-supervised pose (92.9% PCK@20), RuVector embeddings, domain generalization, local SNN, with optional GPU on GCloud.
|
|
argument-hint: "[camera-free|camera-supervised|embeddings|domain-gen|snn|gpu] [--epochs N]"
|
|
---
|
|
|
|
# /ruview-train
|
|
|
|
Train, fine-tune, evaluate, or publish a RuView model.
|
|
|
|
1. Invoke the **`ruview-model-training`** skill.
|
|
2. Pick the track from `$ARGUMENTS`; if empty, ask which:
|
|
- **camera-free** (Track A) — `cargo run -p wifi-densepose-sensing-server -- --pretrain --dataset data/csi/ --pretrain-epochs 50` then `-- --train --dataset data/mmfi/ --epochs 100 --save-rvf model.rvf`. ~84 s on M4 Pro, modest accuracy.
|
|
- **camera-supervised** (Track B, ADR-079) — `python scripts/collect-ground-truth.py`, `python scripts/collect-training-data.py`, `node scripts/align-ground-truth.js`, then train on `data/paired/`, eval with `node scripts/eval-wiflow.js`. ~19 min, 92.9% PCK@20. Needs `data/pose_landmarker_lite.task`.
|
|
- **embeddings** (Track C, AETHER ADR-024) — `wifi-densepose-train` + `wifi-densepose-ruvector`; `-- --model model.rvf --embed`, `-- --build-index env`.
|
|
- **domain-gen** (Track D, MERIDIAN ADR-027) / **snn** (Track E) — `node scripts/snn-csi-processor.js --port 5006`; cognitum-seed-pretraining tutorial.
|
|
- **gpu** — `gcloud config set project cognitum-20260110`; `bash scripts/gcloud-train.sh --gpu l4 --hours 2` (or `--gpu a100 --sweep`, `--dry-run` to smoke-test). VM auto-deletes unless `--keep-vm`.
|
|
3. After training: `cd v2 && cargo test --workspace --no-default-features`, `python archive/v1/data/proof/verify.py`. To publish: `python scripts/publish-huggingface.py`.
|
|
4. Hand off to `/ruview-verify` for the witness bundle.
|