vrr/ruvector
2
0
Fork 0
mirror of https://github.com/ruvnet/RuVector.git synced 2026-07-20 06:33:46 +00:00
Code Issues Projects Releases Packages Wiki Activity Actions 1
ruvector/Cargo.toml
rUv a437ffd034
feat(timesfm): real-model tests + GPU/batch optimization + ruvector-timesfm crate + metaharness (#608) 
* feat(timesfm): GPU/device optimization + ruvector-timesfm integration crate

timesfm:
- cuda/metal features now imply candle (so `--features cuda` alone compiles
  the numeric path); add timesfm::select_device() (TIMESFM_DEVICE=cpu|cuda|metal)
  and use it in the bench instead of hardcoding Device::Cpu.
- Validated real-weight decode on RTX 5080: 45.2 ms (CPU) -> 3.97 ms (cuda) =
  ~11.4x, parity preserved (max-abs 8.58e-6). Note: decode at h<=128 is a single
  forward pass (horizon_len=128), so KV-cache is a no-op there; GPU/f16 are the
  real levers. Derive serde on PruneDecision for the MCP boundary.

ruvector-timesfm (new crate): RuVector-facing integration.
- Forecaster: load-once, forecast(series, horizon) -> point + calibrated p10..p90
  quantile bands.
- anomaly: forecast-band detection (flag observed points outside their p10/p90).
- sweep::EarlyStopper: ADR-191 TimesFM-driven early-stopping for ruflo/Darwin
  sweeps (wraps prune::decide_prune with min_history + confidence gate).
- ruvector-timesfm-forecast: JSON-in/out CLI = the time_series_forecast MCP tool
  entry point.
- telemetry_anomaly example (flags injected spikes on real weights), integration
  tests (5 candle + 3 pure-logic, all green; gated/skip without 814MB weights).

clippy --all-targets -D warnings clean (both feature states); fmt clean.

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

* feat(harness): add generated timesfm metaharness bundle (ADR-041)

Authentic output of the agent-harness-generator (create-agent-harness v0.2.7,
kernel 0.1.2) synthesizing an engineering-pod harness for the TimesFM
forecasting crates. Template vertical:coding (the generator's recommended
rust-crate-harness archetype); host claude-code.

- score: scaffoldReady, 6/6 hard constraints, toolSafety 100, compileConfidence 90
- genome: repo_type rust, topology maintainer/tester/security, risk 0.37,
  mcp_surface local_default_deny
- witness: .harness/manifest.sha256 over .harness/manifest.json, verified valid
  (7c45ab91…). PROVENANCE.md records the repro command, score, genome, witness,
  and the link to the time_series_forecast MCP tool (ruvector-timesfm-forecast).

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

* feat(ruvector-timesfm): batched forecasting (throughput path)

Forecaster::forecast_batch forecasts B equal-length series in one model call.
Measured on real weights (B=32, ctx=256, h=64):
- CPU:  27 -> 166 forecasts/s (6.16x), bit-exact vs per-series
- cuda: 244 -> 2078 forecasts/s (8.45x), rel diff 1.7e-4 (GPU reduction order)

Adds the throughput example (sequential vs batched + correctness check with a
relative tolerance for GPU) and a real-model batch-parity integration test.

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

* feat(harness): Darwin evolve via OpenRouter, key sourced from GCP Secret Manager

Adds scripts/evolve-openrouter.{sh,mjs} to optimize the timesfm-harness with
Darwin Mode's OpenRouter LLM mutator (library-only; not CLI-exposed). The
OpenRouter API key is fetched from GCP Secret Manager at runtime
(gcloud secrets versions access OPENROUTER_API_KEY, project cognitum-20260110)
and exported only into the run's process — never stored in the repo/dotfile/logs.

Driver resolves @metaharness/darwin (devDependency) or DARWIN_DIST for local
monorepo runs. Validated: real-sandbox evolve (1 gen x 2 children,
google/gemini-2.5-flash) scored baseline 0.985 with safety 1.0 and zero
secret-exposure flags; ~$0.003. Mutations pass the validateGeneratedCode gate
and only promote on measured improvement. PROVENANCE.md documents usage.

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

* feat(timesfm): int8/int4 weight quantization (QLinear + load_quantized)

Adds QLinear (full-precision or ggml-quantized weight via QMatMul) threaded
through the decoder; PatchedTimeSeriesDecoder::load_quantized(cfg, vb, dtype)
quantizes the 2 ResidualBlocks + 20 transformer layers (embeddings/norms/scaling
stay f32). Exposed as Forecaster::load_quantized(.., Quant::Q8_0|Q4_0).

Measured on real weights (CPU, ctx=512/h=128) — quant is a MEMORY win, not a
CPU-speed win (dequant overhead dominates the small 16-patch matmuls):
  f32  : 46 ms   814 MB
  Q8_0 : 242 ms  ~212 MB (4x smaller)  rel err 3.5e-3   (recommended)
  Q4_0 : 246 ms  ~112 MB (7x smaller)  rel err 3.1e-2
All outputs finite. f32 path unchanged (QLinear::Full == prior Linear; parity
still 8.58e-6). quant_bench example + Q8_0 integration test added.

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

* feat(ruvector-timesfm): forecast-driven HNSW rebuild scheduler (vector-db hook)

rebuild module: forecast an index's recall-drift curve with TimesFM and advise
WHEN to rebuild — schedule the rebuild to land just before the conservative
(p10) recall forecast crosses a floor, instead of fixed-schedule or
after-the-fact. Forecaster::advise_rebuild(recall_history, floor, horizon,
lead_steps) -> RebuildAdvice{rebuild_now, steps_until_floor, ...}. Ties into the
ruvector-diskann recall-trigger work. Pure-logic + real-model tests.

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

* feat(timesfm): f16-on-load path (Forecaster::load_f16) + GPU bench

Run the forward in f16 (f16 weights/activations). Three localized dtype fixes
make the path f16-clean (attention mask coerce, decode padding dtype, RevIN
scalar-extraction slices); the f32 path is untouched (parity still 8.583e-6).
Forecaster gains a dtype field + load_f16; forecast/forecast_batch build inputs
in the load dtype and surface f32 to callers.

Measured RTX 5080 (B=32, ctx=256, h=64): batched f32 2082 -> f16 3261
forecasts/s (1.57x), sequential 238 -> 303/s. f16 forecasts within rel 2e-2 of
f32. (CPU f16 is slower, like quant — GPU is where f16 pays off.) f16 + Q8
remain the two precision knobs: f16 for GPU latency, Q8_0 for edge memory.

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

---------

Co-authored-by: ruvnet <ruvnet@gmail.com>
2026-06-27 10:48:36 -04:00

373 lines
14 KiB
TOML
Raw Permalink Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

[workspace]
exclude = ["external/ruqu", "external/rvdna", "examples/OSpipe", "examples/rvf", "crates/micro-hnsw-wasm", "crates/ruvector-hyperbolic-hnsw", "crates/ruvector-hyperbolic-hnsw-wasm", "examples/ruvLLM/esp32", "examples/ruvLLM/esp32-flash", "examples/edge-net", "examples/data", "examples/ruvLLM", "examples/delta-behavior", "crates/rvf", "crates/rvf/*", "crates/rvf/*/*", "examples/rvf-desktop", "crates/mcp-brain-server",
# emergent-time-wasm: standalone cdylib with own opt-level=z, panic=abort
"crates/emergent-time-wasm",
# sonic-ct crates: self-contained detached workspaces
"crates/sonic-ct", "crates/sonic-ct-wasm",
# ruvector-postgres is a pgrx-based PostgreSQL extension. Its build script
# requires `$PGRX_HOME` set up via `cargo install cargo-pgrx --version 0.12.9`
# and `cargo pgrx init`, which downloads and builds multiple Postgres
# versions. Keep it out of default workspace builds so `cargo build --workspace`
# works in stock environments. Build it explicitly with `cargo build -p ruvector-postgres`
# after running pgrx init.
"crates/ruvector-postgres",
# Iter 219 (closes ADR-178 Gap E folded into Gap B): the hailo
# crates rejoined the workspace once the iter-218 ruvector-core
# path dep + EmbeddingProvider impls landed. The `hailo` feature
# stays opt-in (only `cargo build --features hailo,cpu-fallback`
# pulls libhailort + candle), so workspace builds on stock x86
# still compile without Pi-specific tooling.
# ruos-thermal: Pi 5 thermal supervisor skeleton (ADR-174). Standalone
# for now; joins workspace once daemon mode + Unix socket protocol
# land in iters 92-97.
"crates/ruos-thermal"]
members = [
"crates/ruvector-temporal-coherence",
"crates/ruvector-acorn",
"crates/ruvector-acorn-wasm",
"crates/ruvector-coherence-hnsw",
"crates/ruvector-rabitq",
"crates/ruvector-rabitq-wasm",
"crates/ruvector-rulake",
"crates/ruvector-core",
"crates/ruvector-node",
"crates/ruvector-wasm",
"crates/ruvector-cli",
"crates/ruvector-bench",
"crates/ruvector-metrics",
"crates/ruvector-filter",
"crates/ruvector-router-core",
"crates/ruvector-router-cli",
"crates/ruvector-router-ffi",
"crates/ruvector-router-wasm",
"crates/ruvector-server",
"crates/ruvector-snapshot",
"crates/ruvector-tiny-dancer-core",
"crates/ruvector-tiny-dancer-wasm",
"crates/ruvector-tiny-dancer-node",
"crates/ruvector-collections",
"crates/ruvector-cluster",
"crates/ruvector-raft",
"crates/ruvector-replication",
"crates/ruvector-graph",
"crates/ruvector-graph-node",
"crates/ruvector-graph-wasm",
"crates/ruvector-gnn",
"crates/ruvector-proof-gate",
"crates/ruvector-gnn-rerank",
"crates/ruvector-gnn-node",
"crates/ruvector-gnn-wasm",
"crates/ruvector-attention",
"crates/ruvector-attention-wasm",
"crates/ruvector-attention-node",
"crates/ruvector-cnn",
"crates/ruvector-cnn-wasm",
"crates/ruvector-mincut",
"crates/ruvector-mincut-wasm",
"crates/ruvector-mincut-node",
"crates/ruvector-mincut-gated-transformer",
"crates/ruvector-mincut-gated-transformer-wasm",
# NOTE: ruvector-postgres is in workspace `exclude` (pgrx env requirement).
"crates/ruvector-nervous-system",
# Iter 219 — hailo backend rejoined the workspace (closes
# ADR-178 Gap E folded into Gap B). All three build clean on
# x86 with default features; opting into the actual NPU path
# requires `--features hailo` on a Pi 5 + AI HAT+.
"crates/hailort-sys",
"crates/ruvector-hailo",
"crates/ruvector-mmwave",
"crates/ruvector-hailo-cluster",
"examples/refrag-pipeline",
"examples/scipix",
"examples/google-cloud",
"examples/subpolynomial-time",
"crates/sona",
"crates/rvlite",
"crates/ruvector-nervous-system",
"crates/ruvector-dag",
"crates/ruvector-dag-wasm",
"crates/ruvector-nervous-system-wasm",
"crates/ruvector-economy-wasm",
"crates/ruvector-learning-wasm",
"crates/ruvector-exotic-wasm",
"crates/ruvector-attention-unified-wasm",
"crates/ruvector-fpga-transformer",
"crates/ruvector-fpga-transformer-wasm",
"crates/ruvector-sparse-inference",
"crates/ruvector-math",
"crates/ruvector-math-wasm",
"examples/benchmarks",
"crates/cognitum-gate-kernel",
"crates/cognitum-gate-tilezero",
"crates/mcp-gate",
"crates/mcp-brain",
"crates/mcp-brain-server",
"crates/ruvllm",
"crates/ruvllm-cli",
"crates/ruvllm-wasm",
"crates/prime-radiant",
"crates/ruvector-delta-core",
"crates/ruvector-delta-wasm",
"crates/ruvector-delta-index",
"crates/ruvector-delta-graph",
"crates/ruvector-delta-consensus",
"crates/ruvector-crv",
"crates/ruvector-temporal-tensor",
"crates/ruvector-domain-expansion",
"crates/ruvector-domain-expansion-wasm",
"crates/ruvector-solver",
"crates/ruvector-solver-wasm",
"crates/ruvector-solver-node",
"crates/ruvector-coherence",
"crates/ruvector-profiler",
"crates/ruvector-attn-mincut",
"crates/ruvector-cognitive-container",
"crates/ruvector-verified",
"crates/ruvector-verified-wasm",
"crates/ruvector-graph-transformer",
"crates/ruvector-graph-transformer-wasm",
"crates/ruvector-graph-transformer-node",
"examples/rvf-kernel-optimized",
"examples/verified-applications",
"crates/thermorust",
"crates/ruvector-dither",
"crates/ruvector-robotics",
"examples/robotics",
"crates/neural-trader-core",
"crates/neural-trader-coherence",
"crates/neural-trader-replay",
"crates/neural-trader-wasm",
# Kalshi integration (ADR-153)
"crates/ruvector-kalshi",
"crates/neural-trader-strategies",
# RuVix Cognition Kernel (organized under crates/ruvix/)
"crates/ruvix/crates/types",
"crates/ruvix/crates/region",
"crates/ruvix/crates/queue",
"crates/ruvix/crates/cap",
"crates/ruvix/crates/proof",
"crates/ruvix/crates/sched",
"crates/ruvix/crates/boot",
"crates/ruvix/crates/vecgraph",
"crates/ruvix/crates/nucleus",
# Phase B: Bare metal AArch64 support
"crates/ruvix/crates/hal",
"crates/ruvix/crates/aarch64",
"crates/ruvix/crates/drivers",
"crates/ruvix/tests",
"crates/ruvix/benches",
"crates/ruvix/examples/cognitive_demo",
# rvAgent — AI Agent Framework (DeepAgents Rust conversion)
"crates/rvAgent/rvagent-core",
"crates/rvAgent/rvagent-backends",
"crates/rvAgent/rvagent-middleware",
"crates/rvAgent/rvagent-tools",
"crates/rvAgent/rvagent-subagents",
"crates/rvAgent/rvagent-cli",
"crates/rvAgent/rvagent-acp",
"crates/rvAgent/rvagent-mcp",
"crates/rvAgent/rvagent-wasm",
"crates/rvAgent/rvagent-a2a",
# ADR-159 a2a-swarm demo
"examples/a2a-swarm",
# ETL pipeline example
"examples/train-discoveries",
# RuView SkyGraph appliance core (ADR-199 Phases 1-4, synthetic ADS-B)
"examples/sky-monitor",
# Browser-facing WASM projection engine for the SkyGraph dashboard (ADR-199 presentation plane)
"examples/sky-monitor/wasm",
# Spectral graph sparsification
"crates/ruvector-sparsifier",
"crates/ruvector-sparsifier-wasm",
# Consciousness metrics (IIT Φ, causal emergence)
"crates/ruvector-consciousness",
"crates/ruvector-consciousness-wasm",
"examples/cmb-consciousness",
"examples/gw-consciousness",
"examples/ecosystem-consciousness",
"examples/quantum-consciousness",
"examples/gene-consciousness",
"examples/climate-consciousness",
# JS bundle decompiler (ADR-135)
"crates/ruvector-decompiler",
"crates/ruvector-decompiler-wasm",
# DiskANN / Vamana (ADR-143)
"crates/ruvector-diskann",
"crates/ruvector-diskann-node",
# Boundary-first scientific discovery PoC
"examples/boundary-discovery",
# CMB Cold Spot boundary-first discovery
"examples/cmb-boundary-discovery",
# FRB population boundary discovery (CHIME-like data)
"examples/frb-boundary-discovery",
# Cosmic void boundary information content
"examples/void-boundary-discovery",
# Multi-regime temporal attractor boundary detection
"examples/temporal-attractor-discovery",
# Music genre boundary discovery via spectral graph bisection
"examples/music-boundary-discovery",
# Weather regime boundary detection (variance/correlation precedes temperature)
"examples/weather-boundary-discovery",
# Market regime boundary discovery via correlation structure
"examples/market-boundary-discovery",
# Health state boundary detection from wearable sensor data
"examples/health-boundary-discovery",
# SETI exotic signals gallery: boundary-first detection of sub-threshold signals
"examples/seti-exotic-signals",
# SETI boundary-first discovery: sub-noise signal detection via coherence graphs
"examples/seti-boundary-discovery",
# Earthquake precursor detection via inter-station correlation boundary shifts
"examples/earthquake-boundary-discovery",
# Pandemic outbreak detection 60 days before case counts via correlation boundaries
"examples/pandemic-boundary-discovery",
# Infrastructure failure prediction via sensor correlation boundaries
"examples/infrastructure-boundary-discovery",
# Pre-seizure detection via brain correlation boundary shifts
"examples/brain-boundary-discovery",
# Clinical-publication-grade pre-seizure detection report with CSV output
"examples/seizure-clinical-report",
# Closed-loop seizure detection + therapeutic response simulation
"examples/seizure-therapeutic-sim",
# Real EEG analysis: CHB-MIT PhysioNet data with boundary-first detection
"examples/real-eeg-analysis",
# Multi-seizure cross-patient analysis: all 7 chb01 seizures
"examples/real-eeg-multi-seizure",
# ruvllm sparse attention kernel for Hailo-10H cluster (ADR-183 – ADR-190)
"crates/ruvllm_sparse_attention",
# Generic retrieval LM + masked discrete diffusion built on the kernel
"crates/ruvllm_retrieval_diffusion",
# RAIRS IVF: Redundant Assignment + Amplified Inverse Residual (ADR-193)
"crates/ruvector-rairs",
# Hybrid sparse-dense search: BM25 + ANN + RRF / RSF / ScoreFusion (ADR-256)
"crates/ruvector-hybrid",
# LSM-ANN: write-optimized streaming vector index for agent memory (ADR-264)
"crates/ruvector-lsm-ann",
# Structure-preserving graph condensation via dynamic min-cut communities
"crates/ruvector-graph-condense",
"crates/ruvector-graph-condense-wasm",
# Perception substrate: delta -> boundary -> coherence -> proof -> action
"crates/ruvector-perception",
# Calculus of emergent / relational time (Wheeler-DeWitt, Page-Wootters,
# entropic, thermal) + Structural Proper Time for agentic systems.
"crates/emergent-time",
# PhotonLayer: learned optical-frontend computing simulator (ADR-260)
"crates/photonlayer-core",
"crates/photonlayer-bench",
"crates/photonlayer-ruvector",
"crates/photonlayer-cli",
"crates/photonlayer-wasm",
# Matryoshka coarse-to-fine ANN search: three-variant funnel with recall/latency tradeoffs (ADR-264)
"crates/ruvector-matryoshka",
# PQ-ADC: Product Quantization with Asymmetric Distance Computation (64× compression)
"crates/ruvector-pq-search",
# SOTA benchmark suite (ADR-265)
"crates/ruvector-sota-bench",
# Capability-gated ANN: per-vector read access control with bitset tokens (ADR-268)
"crates/ruvector-capgated",
# SPANN partition spilling for boundary-safe ANN (ADR-268)
"crates/ruvector-spann",
# ColBERT-style multi-vector MaxSim late-interaction search
"crates/ruvector-maxsim",
# TimesFM 1.0 200M decoder-only patched time-series Transformer (candle, ADR-189/191)
"crates/timesfm",
# RuVector integration for TimesFM: Forecaster + anomaly bands + sweep early-stopping
"crates/ruvector-timesfm",
]
resolver = "2"
[workspace.lints.clippy]
unused_unit = "allow"
[workspace.package]
version = "2.2.3"
edition = "2021"
rust-version = "1.77"
license = "MIT"
authors = ["Ruvector Team"]
repository = "https://github.com/ruvnet/ruvector"
[workspace.dependencies]
# Core functionality
redb = "2.1"
memmap2 = "0.9"
hnsw_rs = "0.3"
simsimd = "5.9"
rayon = "1.10"
crossbeam = "0.8"
# Serialization
rkyv = "0.8"
bincode = { version = "2.0.0-rc.3", features = ["serde"] }
serde = { version = "1.0", features = ["derive"] }
serde_json = "1.0"
# Node.js bindings
napi = { version = "2.16", default-features = false, features = ["napi9", "async", "tokio_rt"] }
napi-derive = "2.16"
# WASM
wasm-bindgen = "0.2"
wasm-bindgen-futures = "0.4"
js-sys = "0.3"
web-sys = { version = "0.3", features = ["Worker", "MessagePort", "console"] }
getrandom = { version = "0.3", features = ["wasm_js"] }
# Async runtime
tokio = { version = "1.41", features = ["rt-multi-thread", "sync", "macros"] }
futures = "0.3"
# Error handling and utilities
thiserror = "2.0"
anyhow = "1.0"
tracing = "0.1"
tracing-subscriber = { version = "0.3", features = ["env-filter"] }
# Math and numerics
nalgebra = { version = "0.33", default-features = false, features = ["std"] }
ndarray = "0.16"
rand = "0.8"
rand_distr = "0.4"
# Time and UUID
chrono = { version = "0.4", features = ["serde"] }
uuid = { version = "1.11", features = ["v4", "serde", "js"] }
# CLI
clap = { version = "4.5", features = ["derive", "cargo"] }
indicatif = "0.17"
console = "0.15"
# Testing and benchmarking
criterion = { version = "0.5", features = ["html_reports"] }
proptest = "1.5"
mockall = "0.13"
# Formal verification
lean-agentic = "=0.1.0"
# Performance
dashmap = "6.1"
parking_lot = "0.12"
once_cell = "1.20"
[profile.release]
opt-level = 3
lto = "fat"
codegen-units = 1
strip = true
panic = "unwind"
[profile.bench]
inherits = "release"
debug = true
[profile.dev]
opt-level = 0
debug = true
[profile.test]
# Patch hnsw_rs to use rand 0.8 instead of 0.9 for WASM compatibility
# This resolves the getrandom version conflict (0.2 vs 0.3)
[patch.crates-io]
hnsw_rs = { path = "./patches/hnsw_rs" }
Reference in a new issue View git blame Copy permalink