iqdb-quantize 1.0.0

Vector quantization (scalar, product, binary) for memory-efficient vector search - part of the iQDB family.
Documentation
# iqdb-quantize v0.1.0 — Scaffold

**The repository bootstrap.** v0.1.0 establishes the structure, tooling, and quality gates the quantization layer is built on — no domain logic yet. It exists so the implementation lands one verified step at a time onto a foundation that already enforces the iQDB engineering standard.

## What is iqdb-quantize?

The memory-efficiency layer of the iQDB vector database. It compresses `f32` embedding vectors into compact codes that preserve similarity-search quality — a million 768-dim vectors drop from ~3 GB to as little as ~96 MB, trading a controlled amount of recall for memory. It ships the three standard schemes (scalar, product, binary) behind one `Quantizer` trait and reuses `iqdb-distance` for distance on the compressed forms. It sits above `iqdb-types` and `iqdb-distance` and is consumed by `iqdb-ivf` for IVF-PQ.

## What's in 0.1.0

- `Cargo.toml` with crate metadata, the Rust 2024 edition, and MSRV `1.87`.
- Dual `Apache-2.0 OR MIT` license files.
- `README.md`, `CHANGELOG.md`, and a documentation skeleton (`docs/API.md`).
- `REPS.md` compliance baseline and the shared lint config (`clippy.toml`, `rustfmt.toml`, `deny.toml`).
- `.github/workflows/ci.yml`: the CI matrix across Linux, macOS, and Windows on stable and the 1.87 MSRV, plus the loom and security (`cargo audit` + `cargo deny`) jobs.
- `dev/DIRECTIVES.md` and `dev/ROADMAP.md`: the committed engineering standards and the phase-by-phase plan to 1.0.

## Breaking changes

**None** — this is the first tag.

## Verification

The scaffold compiles clean and the placeholder smoke test passes across the CI matrix:

```bash
cargo fmt --all -- --check
cargo clippy --all-targets --all-features -- -D warnings
cargo test --all-features
RUSTDOCFLAGS="-D warnings" cargo doc --no-deps --all-features
```

MSRV: Rust 1.87.

## What's next

- **v0.2.0 — scalar quantization (SQ8) + the `Quantizer` trait.** The hard part, not deferred: the first scheme behind the trait every quantizer implements, with per-dimension affine calibration and property-tested round-trip bounds.
- **v0.3.0 — product quantization.** k-means codebooks and asymmetric distance computation (ADC).

## Installation

```toml
[dependencies]
iqdb-quantize = "0.1"
```

## Documentation

- [README](https://github.com/jamesgober/iqdb-quantize/blob/main/README.md)
- [API reference](https://github.com/jamesgober/iqdb-quantize/blob/main/docs/API.md)
- [ROADMAP](https://github.com/jamesgober/iqdb-quantize/blob/main/dev/ROADMAP.md)
- [Standards (REPS)](https://github.com/jamesgober/iqdb-quantize/blob/main/REPS.md)
- [CHANGELOG](https://github.com/jamesgober/iqdb-quantize/blob/main/CHANGELOG.md)

---

**Changelog:** [`CHANGELOG.md`](https://github.com/jamesgober/iqdb-quantize/blob/main/CHANGELOG.md).