Struct Bm25VectorizerBuilder 

pub struct Bm25VectorizerBuilder<TokenIndexer, Tokenizer> { /* private fields */ }

Builder for creating and configuring a Bm25Vectorizer.

It supports fitting on a corpus to automatically compute the average document length, and validates all parameters before building.

Type Parameters

• TokenIndexer: Implementation of Bm25TokenIndexer trait
• Tokenizer: Implementation of Bm25Tokenizer trait

Examples

Basic usage with manual avgdl:

use bm25_vectorizer::{Bm25VectorizerBuilder, MockWhitespaceTokenizer, MockHashTokenIndexer};

let vectorizer = Bm25VectorizerBuilder::new()
    .tokenizer(MockWhitespaceTokenizer)
    .token_indexer(MockHashTokenIndexer)
    .k1(1.2)
    .b(0.75)
    .avgdl(10.0)
    .build()?;

Usage with corpus fitting:

use bm25_vectorizer::{Bm25VectorizerBuilder, MockWhitespaceTokenizer, MockHashTokenIndexer};

let corpus = vec!["hello world", "world of rust", "hello rust programming"];
let vectorizer = Bm25VectorizerBuilder::new()
    .tokenizer(MockWhitespaceTokenizer)
    .token_indexer(MockHashTokenIndexer)
    .k1(1.2)
    .b(0.75)
    .fit(&corpus)?  // Automatically computes avgdl
    .build()?;

Implementations

impl<TokenIndexer, Tokenizer> Bm25VectorizerBuilder<TokenIndexer, Tokenizer>

pub fn new() -> Self

pub fn k1(self, k1: f32) -> Self

pub fn b(self, b: f32) -> Self

pub fn delta(self, delta: f32) -> Self

pub fn avgdl(self, avgdl: f32) -> Self

pub fn tokenizer(self, tokenizer: Tokenizer) -> Self

pub fn token_indexer(self, token_indexer: TokenIndexer) -> Self

pub fn fit(self, corpus: &[&str]) -> Result<Self, Bm25VectorizerError>

where Tokenizer: Bm25Tokenizer + Sync,

pub fn fit_iter<I, S>(self, corpus: I) -> Result<Self, Bm25VectorizerError>

where I: IntoIterator<Item = S>, S: AsRef<str>, Tokenizer: Bm25Tokenizer + Sync,

pub fn build( self, ) -> Result<Bm25Vectorizer<TokenIndexer, Tokenizer>, Bm25VectorizerError>