Struct CountVectorizer 

pub struct CountVectorizer { /* private fields */ }

Converts text documents into a sparse term-count matrix.

Each document becomes a row, each unique token a column. Cell values are the number of times that token appears in that document.

Example

use scry_learn::text::CountVectorizer;

let mut cv = CountVectorizer::new();
let docs = ["the cat sat", "the dog sat", "the cat played"];
let matrix = cv.fit_transform(&docs);

assert_eq!(matrix.n_rows(), 3);
assert_eq!(matrix.n_cols(), cv.vocabulary().len());

Implementations

impl CountVectorizer

pub fn new() -> Self

Create a new CountVectorizer with default settings.

pub fn min_df(self, n: usize) -> Self

Set minimum document frequency (absolute). Tokens appearing in fewer documents are excluded. Default: 1.

pub fn max_df(self, frac: f64) -> Self

Set maximum document frequency as a fraction in (0.0, 1.0]. Tokens appearing in more than this fraction of documents are excluded. Default: 1.0 (no filtering).

pub fn ngram_range(self, min_n: usize, max_n: usize) -> Self

Set n-gram range. Default: (1, 1) (unigrams only).

pub fn max_features(self, n: usize) -> Self

Limit vocabulary to the top n features by total frequency. Default: no limit.

pub fn binary(self, b: bool) -> Self

If true, all non-zero counts become 1 (presence/absence). Default: false.

pub fn fit<S: AsRef<str>>(&mut self, documents: &[S])

Learn vocabulary from documents.

pub fn transform<S: AsRef<str>>(&self, documents: &[S]) -> CsrMatrix

Transform documents into a sparse CSR matrix of counts.

Panics if fit() has not been called.

pub fn fit_transform<S: AsRef<str>>(&mut self, documents: &[S]) -> CsrMatrix

Fit the vocabulary and transform in one step.

pub fn vocabulary(&self) -> &HashMap<String, usize>

Return the learned vocabulary (token → column index).

pub fn get_feature_names(&self) -> Vec<String>

Return feature names sorted by column index.

pub fn n_features(&self) -> usize

Number of features in the vocabulary.

pub fn is_fitted(&self) -> bool

Whether fit() has been called.

Trait Implementations

impl Clone for CountVectorizer

fn clone(&self) -> CountVectorizer

Returns a duplicate of the value.

fn clone_from(&mut self, source: &Self)

Performs copy-assignment from source.

impl Debug for CountVectorizer

fn fmt(&self, f: &mut Formatter<'_>) -> Result

Formats the value using the given formatter.

impl Default for CountVectorizer

fn default() -> Self

Returns the “default value” for a type.