Struct TfIdfLogisticClassifier 

pub struct TfIdfLogisticClassifier { /* private fields */ }

Logistic regression classifier trained on TF-IDF features.

Uses the existing TfidfVectorizer to compute features and mini-batch stochastic gradient descent (SGD) to train the model.

Example

use scirs2_text::text_classification::TfIdfLogisticClassifier;

let mut clf = TfIdfLogisticClassifier::new();
let texts  = &["good great excellent", "bad terrible awful", "okay decent fine", "poor mediocre subpar"];
let labels = &["pos", "neg", "pos", "neg"];
clf.fit(texts, labels, 50, 0.1).unwrap();
let pred = clf.predict("excellent wonderful").unwrap();
assert!(!pred.is_empty());

Implementations

impl TfIdfLogisticClassifier

pub fn new() -> Self

Create a new classifier with default TF-IDF settings.

pub fn fit( &mut self, texts: &[&str], labels: &[&str], max_iter: usize, lr: f64, ) -> Result<()>

Fit the classifier.

Parameters

• texts: Training documents.
• labels: Corresponding class labels.
• max_iter: Number of SGD epochs.
• lr: Learning rate for SGD.

Errors

Returns TextError::InvalidInput when inputs are inconsistent.

pub fn predict(&self, text: &str) -> Result<String>

Predict the most probable class for text.

Errors

Returns TextError::ModelNotFitted when the model has not been trained.

pub fn predict_proba(&self, text: &str) -> Result<Vec<f64>>

Return the probability distribution over classes for text.

Errors

Returns TextError::ModelNotFitted when the model has not been trained.

Trait Implementations

impl Default for TfIdfLogisticClassifier

fn default() -> Self

Returns the “default value” for a type.