ruview @ v1024

integrity

size: 55.1 MiB
downloaded: 2026-07-23 14:07 UTC
last checked: 2026-07-23 14:07 UTC

source https://github.com/ruvnet/RuView · available · github

release notes

Automated release from CI pipeline

Changes: feat(cog-person-count): train count_v1.safetensors — honest v0.0.1 (ADR-103) (#695)

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

Docker Image: ghcr.io/ruvnet/RuView:6b4994e1052873e4249a5fbed7db54878e319e2d

download

unix · zip

curl -fL -o v1024.zip https://ratatoskr.space/pkg/ruview/v1024.zip
                    printf '%s  %s\n' '30d8f58dbd44faba6ad218755f59123efb1b012d55f3f20007ddbc95a58aee44' 'v1024.zip' | sha256sum -c -

windows · zip

$url = "https://ratatoskr.space/pkg/ruview/v1024.zip"
$out = "v1024.zip"
Invoke-WebRequest -Uri $url -OutFile $out
if ((Get-FileHash $out -Algorithm SHA256).Hash.ToLowerInvariant() -ne "30d8f58dbd44faba6ad218755f59123efb1b012d55f3f20007ddbc95a58aee44") { throw "sha256 mismatch" }

unix · tar.gz

curl -fL -o v1024.tar.gz https://ratatoskr.space/pkg/ruview/v1024.tar.gz
                    printf '%s  %s\n' '357339e4f740afa6d46009cbd2047bb40a80bb96b28a8dcd0c39cf56ac00d044' 'v1024.tar.gz' | sha256sum -c -

windows · tar.gz

$url = "https://ratatoskr.space/pkg/ruview/v1024.tar.gz"
$out = "v1024.tar.gz"
Invoke-WebRequest -Uri $url -OutFile $out
if ((Get-FileHash $out -Algorithm SHA256).Hash.ToLowerInvariant() -ne "357339e4f740afa6d46009cbd2047bb40a80bb96b28a8dcd0c39cf56ac00d044") { throw "sha256 mismatch" }

download via yggdrasil mesh

unix · zip

curl -fL -o v1024.zip http://14cc7d57b5e70f679b851fe5b272ce17c70632ff4beb5b35ab64bc706b2485af.pk.ygg/pkg/ruview/v1024.zip
                    printf '%s  %s\n' '30d8f58dbd44faba6ad218755f59123efb1b012d55f3f20007ddbc95a58aee44' 'v1024.zip' | sha256sum -c -

windows · zip

$url = "http://14cc7d57b5e70f679b851fe5b272ce17c70632ff4beb5b35ab64bc706b2485af.pk.ygg/pkg/ruview/v1024.zip"
$out = "v1024.zip"
Invoke-WebRequest -Uri $url -OutFile $out
if ((Get-FileHash $out -Algorithm SHA256).Hash.ToLowerInvariant() -ne "30d8f58dbd44faba6ad218755f59123efb1b012d55f3f20007ddbc95a58aee44") { throw "sha256 mismatch" }

unix · tar.gz

curl -fL -o v1024.tar.gz http://14cc7d57b5e70f679b851fe5b272ce17c70632ff4beb5b35ab64bc706b2485af.pk.ygg/pkg/ruview/v1024.tar.gz
                    printf '%s  %s\n' '357339e4f740afa6d46009cbd2047bb40a80bb96b28a8dcd0c39cf56ac00d044' 'v1024.tar.gz' | sha256sum -c -

windows · tar.gz

$url = "http://14cc7d57b5e70f679b851fe5b272ce17c70632ff4beb5b35ab64bc706b2485af.pk.ygg/pkg/ruview/v1024.tar.gz"
$out = "v1024.tar.gz"
Invoke-WebRequest -Uri $url -OutFile $out
if ((Get-FileHash $out -Algorithm SHA256).Hash.ToLowerInvariant() -ne "357339e4f740afa6d46009cbd2047bb40a80bb96b28a8dcd0c39cf56ac00d044") { throw "sha256 mismatch" }

artifact	format	size	hashes
v1024.zip	zip	55.1 MiB	blake3-24 49f0b68e171fd5fbe61b8cbf5e1203add7c629b330adda31 sha256 30d8f58dbd44faba6ad218755f59123efb1b012d55f3f20007ddbc95a58aee44 sha1 35bba57ba32e1fff8f6ae6e27a8c3eaa94735cc2
v1024.tar.gz	tar.gz	53.7 MiB	blake3-24 927839b2ad626ec59254b8de69257eff1a5156ad57ec90d7 sha256 357339e4f740afa6d46009cbd2047bb40a80bb96b28a8dcd0c39cf56ac00d044 sha1 9b010b3ad48df6ab4d8ca28249f5398bcbfd0666

install

bazel

http_archive(
    name = "ruview",
    urls = ["https://ratatoskr.space/pkg/ruview/v1024.tar.gz"],
    integrity = "sha256-NXM55PdAr6bUYAnL0gR7tAqAu5ayio3NDDnPVqwA0EQ=",
    strip_prefix = "ruview-v1024",
)

zig

.url = "https://ratatoskr.space/pkg/ruview/v1024.tar.gz",

install via yggdrasil mesh

bazel

http_archive(
    name = "ruview",
    urls = ["http://14cc7d57b5e70f679b851fe5b272ce17c70632ff4beb5b35ab64bc706b2485af.pk.ygg/pkg/ruview/v1024.tar.gz"],
    integrity = "sha256-NXM55PdAr6bUYAnL0gR7tAqAu5ayio3NDDnPVqwA0EQ=",
    strip_prefix = "ruview-v1024",
)

zig

.url = "http://14cc7d57b5e70f679b851fe5b272ce17c70632ff4beb5b35ab64bc706b2485af.pk.ygg/pkg/ruview/v1024.tar.gz",

← v1027 v1021 →