Expand description
Lategrep: CPU-based PLAID implementation for multi-vector search
This crate provides a pure Rust, CPU-only implementation of the PLAID algorithm for efficient multi-vector search (late interaction retrieval).
Re-exports§
pub use codec::CentroidStore;pub use codec::ResidualCodec;pub use error::Error;pub use error::Result;pub use index::Index;pub use index::IndexConfig;pub use index::LoadedIndex;pub use index::Metadata;pub use kmeans::compute_centroids;pub use kmeans::compute_centroids_from_documents;pub use kmeans::compute_kmeans;pub use kmeans::estimate_num_partitions;pub use kmeans::ComputeKmeansConfig;pub use search::QueryResult;pub use search::SearchParameters;pub use strided_tensor::IvfStridedTensor;pub use strided_tensor::StridedTensor;pub use update::UpdateConfig;
Modules§
- codec
- Residual codec for quantization and decompression
- delete
- Document deletion functionality for removing documents from an existing index.
- embeddings
- Embedding reconstruction from compressed index data.
- error
- Error types for the next-plaid library
- index
- Index creation and management for PLAID
- kmeans
- K-means clustering integration using fastkmeans-rs.
- mmap
- Memory-mapped file support for efficient large index loading.
- search
- Search functionality for PLAID
- strided_
tensor - StridedTensor for efficient batch lookup of variable-length sequences.
- update
- Index update functionality for adding new documents.
- utils
- Utility functions for next-plaid
Structs§
- FastK
Means - Fast k-means clustering implementation compatible with ndarray.
- KMeans
Config - Configuration for the FastKMeans algorithm