mirror of
https://github.com/ruvnet/RuView.git
synced 2026-07-10 01:30:01 +00:00
6b4994e105
Phase 2 of ADR-103: trained count head on the existing 1,077 paired samples (the same data that produced pose_v1 yesterday). Honest result: 65.1% eval accuracy / 100% within ±1 / MAE 0.349 on the held-out time-window. Per-class: 100% on "empty room" / 0% on "1 person". The model overfit by epoch 100 (train_acc → 1.0, eval_loss climbed 0.67 → 7.8) and the "best" checkpoint is the snapshot that happened to predict the eval window's class distribution (140/215 = 65.1%, matches eval_acc exactly). Confidence head Spearman = 0.023 ⇒ uncalibrated. Same data-bound failure mode as pose_v1 (#645), bounded by single-session training data; same fix path (multi-room). What v0.0.1 still validates end-to-end: * PyTorch → safetensors → Candle Rust loads cleanly on first try. `cog-person-count health` reports `backend: candle-cpu` and emits real per-frame predictions instead of the stub backend's hard-coded {1 person, 0 confidence}. Architecture parity between train-count.py and src/inference.rs::CountNet is bit-exact. * ONNX export bit-clean (16 KB, opset 18, dynamic batch axis). * Training wall time: 5.6 s for 400 epochs on RTX 5080. * Binary size unchanged (2.36 MB stripped), model loads via mmap at runtime. This commit ships: * scripts/align-ground-truth.js: extended to emit n_persons_mode + n_persons_max per window so the training pipeline has count labels. Backwards-compatible (additive fields). * scripts/train-count.py: new — mirrors CountNet architecture exactly, loads paired.jsonl, trains 400 epochs with CE+BCE+Brier loss, exports safetensors + ONNX + per-epoch JSON. * v2/.../cog/artifacts/{count_v1.safetensors,count_v1.onnx, count_train_results.json}: the trained artifacts. * v2/.../cog/README.md: Status table updated with the v0.0.1 numbers + an Honest Caveat section explaining the data-bound result. * docs/benchmarks/person-count-cog.md: new — full v0.0.1 benchmark log mirroring the format docs/benchmarks/pose-estimation-cog.md established. Includes comparison to ADR-103 v0.1.0 acceptance gates and per-class breakdown. Still pending: * `run` subcommand wiring (long-running polling loop, same as pose) * Cross-compile + sign + GCS upload (mirror of pose cog pipeline) * Live install on cognitum-v0 * v0.2.0: re-train on multi-room data, LoRA per-room adapters, Stoer-Wagner min-cut clip in fusion stage |
||
|---|---|---|
| .. | ||
| swarm_presets | ||
| align-ground-truth.js | ||
| apnea-detector.js | ||
| benchmark-model.py | ||
| benchmark-rf-scan.js | ||
| benchmark-ruvllm.js | ||
| benchmark-wiflow.js | ||
| check_fix_markers.py | ||
| check_health.py | ||
| collect-ground-truth.py | ||
| collect-training-data.py | ||
| csi-graph-visualizer.js | ||
| csi-spectrogram.js | ||
| deep-scan.js | ||
| device-fingerprint.js | ||
| esp32_jsonl_to_rvcsi.py | ||
| esp32_wasm_test.py | ||
| eval-wiflow.js | ||
| export-onnx.py | ||
| fix-markers.json | ||
| gait-analyzer.js | ||
| gcloud-train.sh | ||
| generate-witness-bundle.sh | ||
| generate_nvs_matrix.py | ||
| inject_fault.py | ||
| install-qemu.sh | ||
| mac-mini-train.sh | ||
| material-classifier.js | ||
| material-detector.js | ||
| mesh-graph-transformer.js | ||
| mincut-person-counter.js | ||
| mmwave_fusion_bridge.py | ||
| passive-radar.js | ||
| probe-fft-platform.py | ||
| provision.py | ||
| publish-huggingface.py | ||
| publish-huggingface.sh | ||
| qemu-chaos-test.sh | ||
| qemu-cli.sh | ||
| qemu-esp32s3-test.sh | ||
| qemu-mesh-test.sh | ||
| qemu-snapshot-test.sh | ||
| qemu_swarm.py | ||
| record-csi-udp.py | ||
| release-v0.5.4.sh | ||
| rf-scan-multifreq.js | ||
| rf-scan.js | ||
| rf-tomography.js | ||
| room-fingerprint.js | ||
| seed_csi_bridge.py | ||
| sleep-monitor.js | ||
| snn-csi-processor.js | ||
| stress-monitor.js | ||
| swarm_health.py | ||
| through-wall-detector.js | ||
| train-camera-free.js | ||
| train-count.py | ||
| train-ruvllm.js | ||
| train-wiflow-supervised.js | ||
| train-wiflow.js | ||
| training-config-sweep.json | ||
| udp-relay.py | ||
| validate_mesh_test.py | ||
| validate_qemu_output.py | ||
| wiflow-model.js | ||