Struct FibQuantAdapter 

pub struct FibQuantAdapter { /* private fields */ }

Adapter for the fib-quant crate (k=4, N=32, paper core path).

Implementations

impl FibQuantAdapter

pub fn new( dim: usize, k: u32, n: u32, training_samples: u32, lloyd_restarts: u32, lloyd_iterations: u32, ) -> Result<Self>

pub fn build_quantizer(&self, seed: u64) -> Result<FibQuantizer, PolyKvError>

Build a FibQuantizer for the given seed.

pub fn decode_codes_payload( &self, payload: &[u8], seed: u64, ) -> Result<Vec<FibCodeV1>>

Decode one or more FibCodeV1 values from a pool/shell payload without reconstructing the f32 vector values. This is the cold-pool compressed read path used by query-aware attention selection.

Trait Implementations

impl KVecCodec for FibQuantAdapter

Available on crate feature fib only.

fn codec_id(&self) -> CodecId

Return the codec identifier (“fib_k4_n32”, “turbo_8bit”, etc.).

fn encode(&self, vector: &[f32], seed: u64) -> Result<Vec<u8>>

Encode a vector of f32 values into a compressed byte payload.

fn encode_batch(&self, vectors: &[&[f32]], seed: u64) -> Result<Vec<Vec<u8>>>

Encode a batch of vectors in one call.

fn encode_batch_compact( &self, vectors: &[&[f32]], seed: u64, ) -> Result<Option<Vec<u8>>>

Encode a batch into one amortized binary payload when supported.

fn decode(&self, payload: &[u8], seed: u64) -> Result<Vec<f32>>

Decode a compressed byte payload back into a vector of f32 values.

fn decode_batch(&self, payloads: &[&[u8]], seed: u64) -> Result<Vec<Vec<f32>>>

Decode a batch of compressed payloads. Default loops over decode.

fn decode_batch_compact( &self, payload: &[u8], seed: u64, ) -> Result<Option<Vec<Vec<f32>>>>

Decode one compact batch payload back into vectors when supported.

fn dim(&self) -> usize

The expected dimension of input/output vectors.

fn compression_ratio(&self) -> f64

Expected compression ratio (nominal).

fn is_gpu_accelerated(&self) -> bool

True if this adapter has access to GPU acceleration at runtime.

fn is_gpu_accelerated_for(&self, n: usize, d: usize) -> bool

True if a batch of n vectors at dimension d would actually dispatch to GPU. Default falls back to the device-availability probe; codecs with a runtime threshold should override.

fn codebook_digest(&self, seed: u64) -> Option<String>

Codebook digest (BLAKE3 hex) for provenance receipts. Returns None if the codec doesn’t support this.

fn rotation_digest(&self, seed: u64) -> Option<String>

Rotation digest (BLAKE3 hex) for provenance receipts. Returns None if the codec doesn’t support this.