Ivf Flat Index Params Module

Rust module: cuvs::ivf_flat::index_params

Source: rust/cuvs/src/ivf_flat/index_params.rs

IndexParams

1
pub struct IndexParams(pub ffi::cuvsIvfFlatIndexParams_t);

Methods

new

1
pub fn new() -> Result<IndexParams> { ... }

Returns a new IndexParams

Source: rust/cuvs/src/ivf_flat/index_params.rs:15

set_n_lists

1
pub fn set_n_lists(self, n_lists: u32) -> IndexParams { ... }

The number of clusters used in the coarse quantizer.

Source: rust/cuvs/src/ivf_flat/index_params.rs:24

set_metric

1
pub fn set_metric(self, metric: DistanceType) -> IndexParams { ... }

DistanceType to use for building the index

Source: rust/cuvs/src/ivf_flat/index_params.rs:32

set_metric_arg

1
pub fn set_metric_arg(self, metric_arg: f32) -> IndexParams { ... }

The number of iterations searching for kmeans centers during index building.

Source: rust/cuvs/src/ivf_flat/index_params.rs:40

set_kmeans_n_iters

1
pub fn set_kmeans_n_iters(self, kmeans_n_iters: u32) -> IndexParams { ... }

The number of iterations searching for kmeans centers during index building.

Source: rust/cuvs/src/ivf_flat/index_params.rs:47

set_kmeans_trainset_fraction

1
pub fn set_kmeans_trainset_fraction(self, kmeans_trainset_fraction: f64) -> IndexParams { ... }

If kmeans_trainset_fraction is less than 1, then the dataset is subsampled, and only n_samples * kmeans_trainset_fraction rows are used for training.

Source: rust/cuvs/src/ivf_flat/index_params.rs:57

set_add_data_on_build

1
pub fn set_add_data_on_build(self, add_data_on_build: bool) -> IndexParams { ... }

After training the coarse and fine quantizers, we will populate the index with the dataset if add_data_on_build == true, otherwise the index is left empty, and the extend method can be used to add new vectors to the index.

Source: rust/cuvs/src/ivf_flat/index_params.rs:68

Source: rust/cuvs/src/ivf_flat/index_params.rs:11