Trait Predict 

pub trait Predict<T>
where
    Self: Sync,
{
    // Required methods
    fn predict_value(&self, problem: &mut FeatureVector<T>) -> Result<(), Error>;
    fn predict_probability(
        &self,
        problem: &mut FeatureVector<T>,
    ) -> Result<(), Error>;
}

Implemented by DenseSVM and SparseSVM to predict a FeatureVector.

Predicting a label

To predict a label, first make sure the FeatureVector has all features set. Then calling

use ffsvm::{DenseFeatures, DenseSVM, Predict};

fn set_features(svm: &DenseSVM, problem: &mut DenseFeatures) {
    // Predicts the value.
    svm.predict_value(problem);
}

will update the FeatureVector::label to correspond to the class label with the highest likelihood.

Predicting a label and obtaining probability estimates.

If the libSVM model was trained with probability estimates FFSVM can not only predict the label, but it can also give information about the likelihood distribution of all classes. This can be helpful if you want to consider alternatives.

Probabilities are estimated like this:

use ffsvm::{DenseFeatures, DenseSVM, Predict};

fn set_features(svm: &DenseSVM, features: &mut DenseFeatures) {
    // Predicts the value.
    svm.predict_probability(features);
}

Predicting probabilities automatically predicts the best label. In addition, FeatureVector::probabilities will be updated accordingly. The class labels for each probablity entry can be obtained by the SVM’s class_label_for_index and class_index_for_label methods.

Required Methods

fn predict_value(&self, problem: &mut FeatureVector<T>) -> Result<(), Error>

Predict a single value for a FeatureVector.

The problem needs to have all features set. Once this method returns, the FeatureVector::label will be set.

Errors

Can fail if the feature vector didn’t match the original SVM configuration.

fn predict_probability( &self, problem: &mut FeatureVector<T>, ) -> Result<(), Error>

Predict a probability value for a problem.

The problem needs to have all features set. Once this method returns, both FeatureVector::label will be set, and all FeatureVector::probabilities will be available accordingly.

Errors

Can fail if the model was not trained with probability estimates support.

Implementors

impl Predict<VecSimd<f32x8>> for DenseSVM