ruv 8f2de7e9f2 feat: ADR-072 WiFlow SOTA architecture — TCN + axial attention + pose decoder ...

Pure JS implementation of WiFlow (arXiv:2602.08661) adapted for ESP32:
- TCN temporal encoder (dilated causal conv, k=7, dilation 1/2/4/8)
- Asymmetric spatial encoder (1x3 residual blocks, stride-2)
- Axial self-attention (width + height, 8 heads, 256 channels)
- Pose decoder (adaptive pooling → 17x2 COCO keypoints)
- SmoothL1 + bone constraint loss (14 skeleton connections)
- 1.8M params (1.6 MB at INT8), 198M FLOPs

Integrated with camera-free pipeline (pose proxy labels from
RSSI triangulation + subcarrier asymmetry + vibration)

Co-Authored-By: claude-flow <ruv@ruv.net>

2026-04-02 23:40:23 -04:00

..

swarm_presets

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

benchmark-model.py

feat: GCloud GPU training pipeline + data collection + benchmarking

2026-04-02 22:04:57 -04:00

benchmark-ruvllm.js

fix: ruvllm pipeline — 7 critical fixes, all metrics improved

2026-04-02 22:40:48 -04:00

benchmark-wiflow.js

feat: ADR-072 WiFlow SOTA architecture — TCN + axial attention + pose decoder

2026-04-02 23:40:23 -04:00

check_health.py

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

collect-training-data.py

feat: GCloud GPU training pipeline + data collection + benchmarking

2026-04-02 22:04:57 -04:00

esp32_wasm_test.py

feat: add ADR-042 CHCI protocol, 24 new edge modules, README restructure

2026-03-03 11:35:57 -05:00

gcloud-train.sh

feat: GCloud GPU training pipeline + data collection + benchmarking

2026-04-02 22:04:57 -04:00

generate_nvs_matrix.py

fix(firmware): fall detection, 4MB flash, QEMU CI (#263, #265)

2026-03-15 11:49:29 -04:00

generate-witness-bundle.sh

feat: 100% validated witness bundle with proof hash + generator script

2026-03-01 15:51:38 -05:00

inject_fault.py

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

install-qemu.sh

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

mac-mini-train.sh

feat: Mac Mini M4 Pro training script (7-step pipeline)

2026-04-02 22:42:32 -04:00

mmwave_fusion_bridge.py

feat: ADR-063/064 mmWave sensor fusion + multimodal ambient intelligence (#269)

2026-03-15 16:10:10 -04:00

provision.py

fix(firmware): provision.py nvs import + partition config template

2026-03-10 08:40:47 -04:00

publish-huggingface.py

feat: HuggingFace model publishing pipeline + model card

2026-04-02 22:04:16 -04:00

publish-huggingface.sh

feat: HuggingFace model publishing pipeline + model card

2026-04-02 22:04:16 -04:00

qemu_swarm.py

fix(firmware): fall detection, 4MB flash, QEMU CI (#263, #265)

2026-03-15 11:49:29 -04:00

qemu-chaos-test.sh

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

qemu-cli.sh

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

qemu-esp32s3-test.sh

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

qemu-mesh-test.sh

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

qemu-snapshot-test.sh

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

release-v0.5.4.sh

feat: ADR-069 ESP32 CSI → Cognitum Seed RVF pipeline (v0.5.4-esp32)

2026-04-02 19:32:18 -04:00

seed_csi_bridge.py

feat: ADR-069 ESP32 CSI → Cognitum Seed RVF pipeline (v0.5.4-esp32)

2026-04-02 19:32:18 -04:00

swarm_health.py

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

train-camera-free.js

feat: camera-free 17-keypoint pose training (10 sensor signals)

2026-04-02 23:05:07 -04:00

train-ruvllm.js

fix: ruvllm pipeline — 7 critical fixes, all metrics improved

2026-04-02 22:40:48 -04:00

train-wiflow.js

feat: ADR-072 WiFlow SOTA architecture — TCN + axial attention + pose decoder

2026-04-02 23:40:23 -04:00

training-config-sweep.json

feat: GCloud GPU training pipeline + data collection + benchmarking

2026-04-02 22:04:57 -04:00

validate_mesh_test.py

feat: QEMU ESP32-S3 testing platform + swarm configurator (ADR-061/062) (#260)

2026-03-14 13:39:51 -04:00

validate_qemu_output.py

fix(firmware): fall detection, 4MB flash, QEMU CI (#263, #265)

2026-03-15 11:49:29 -04:00

wiflow-model.js

feat: ADR-072 WiFlow SOTA architecture — TCN + axial attention + pose decoder

2026-04-02 23:40:23 -04:00