mirror of
https://github.com/ruvnet/RuView.git
synced 2026-07-09 17:19:29 +00:00
1f05456588
Some checks failed
Continuous Deployment / Notify Deployment Status (push) Blocked by required conditions
Continuous Deployment / Post-deployment Monitoring (push) Blocked by required conditions
Continuous Deployment / Pre-deployment Checks (push) Waiting to run
Continuous Deployment / Deploy to Staging (push) Blocked by required conditions
Continuous Deployment / Deploy to Production (push) Blocked by required conditions
Continuous Deployment / Rollback Deployment (push) Blocked by required conditions
Fix-Marker Regression Guard / Verify fix markers (push) Waiting to run
Security Scanning / Secret Scanning (push) Waiting to run
Security Scanning / Container Security Scan (push) Waiting to run
Security Scanning / Infrastructure Security Scan (push) Waiting to run
Security Scanning / Dependency Vulnerability Scan (push) Waiting to run
Security Scanning / Static Application Security Testing (push) Waiting to run
Security Scanning / License Compliance Scan (push) Waiting to run
Security Scanning / Security Policy Compliance (push) Waiting to run
Security Scanning / Security Report (push) Blocked by required conditions
Continuous Integration / Tests-2 (push) Has been cancelled
Continuous Integration / Code Quality & Security (push) Has been cancelled
Continuous Integration / Rust Workspace Tests (push) Has been cancelled
Continuous Integration / Tests (push) Has been cancelled
Continuous Integration / Tests-1 (push) Has been cancelled
Continuous Integration / Performance Tests (push) Has been cancelled
Continuous Integration / Docker Build & Test (push) Has been cancelled
Continuous Integration / API Documentation (push) Has been cancelled
Continuous Integration / Notify (push) Has been cancelled
* feat(ADR-261): multi-bit (b∈{1,2,4}) quantized HNSW traversal + scaling harness
Generalize the SymphonyQG-style quantized-traversal HNSW from 1-bit Hamming to a
b-bit-per-dimension code (b ∈ {1,2,4}), mirroring ADR-156 §10's multi-bit RaBitQ
scheme (rotate via FHT Pass-2, uniform mid-rise scalar quantizer over [-3,3],
ranked by per-dim L1). b=1 is byte-for-byte the original construction (codes in
{0,1} ⇒ L1 == Hamming), pinned by one_bit_build_bits_matches_legacy_build.
Bytes/node scales linearly: 128-d → 16/32/64 B for b=1/2/4.
- hnsw_quantized.rs: QuantizedHnswIndex::build_bits(...,bits,...), bits()/
bytes_per_node() accessors, code-L1 greedy+beam traversal. build(...) kept as
the b=1 backward-compatible entry point. +4 tests (multi-bit recall regression,
bits clamp, bytes/node, legacy parity).
- ann_measure.rs: build_indices_bits / build_quant_bits / run_scaling_study +
best_float_op / best_quant_op; scaling_report (#[ignore], --release) and a
CI-safe scaling_study_small_is_consistent.
- ann_bench.rs: 2-bit and 4-bit quant criterion benches over the shared graph.
ruvector lib 151 → 156 passed, 0 failed, 1 ignored (scaling_report).
Co-Authored-By: claude-flow <ruv@ruv.net>
* docs(adr-261): record M2 multi-bit scaling study — measured, no crossover (refutes M1 prediction)
Multi-bit (b∈{1,2,4}) quantized HNSW traversal + N∈{10k,100k,250k} scaling study,
measured on this box. No crossover at any (N,b): at 10k more bits help (ratio
0.19→0.48×, b≥2 reaches 0.90 recall) but quant stays slower than float HNSW at
equal recall; at 100k/250k quant recall collapses (b=4: 1.0→0.788→0.624, never
≥0.90) while float holds ≥0.92. The predicted large-N crossover moved the wrong
way. Published negative with the mechanism explained. ADR-261 §11.
Co-Authored-By: claude-flow <ruv@ruv.net>
---------
Co-authored-by: ruv <ruvnet@gmail.com>
|
||
|---|---|---|
| .. | ||
| .cargo | ||
| .claude-flow | ||
| crates | ||
| data | ||
| docs | ||
| examples | ||
| patches/ruvector-crv | ||
| .gitignore | ||
| Cargo.lock | ||
| Cargo.toml | ||
| rust-toolchain.toml | ||
| ruvector.db | ||