Struct ppca::InferredMaskedMix

pub struct InferredMaskedMix { /* private fields */ }

The inferred probability distribution in the state space of a given sample of a PPCA Mixture Model. This class is the analogous of InferredMasked for the PPCAMix model.

Implementations

impl InferredMaskedMix

pub fn log_posterior(&self) -> &DVector<f64>

The logarithm of the posterior distribution over the PPCA model indices.

pub fn posterior(&self) -> DVector<f64>

The posterior distribution over the PPCA model indices.

pub fn state(&self) -> DVector<f64>

The mean of the posterior distribution in the state space.

pub fn covariance(&self) -> DMatrix<f64>

The covariance matrices of the posterion distribution in the state space.

pub fn smoothed(&self, mix: &PPCAMix) -> DVector<f64>

The smoothed output values for a given output model.

pub fn extrapolated(&self, mix: &PPCAMix, sample: &MaskedSample) -> DVector<f64>

The extrapolated output values for a given output model and the corresponding sample.

pub fn smoothed_covariance(&self, mix: &PPCAMix) -> DMatrix<f64>

The covariance for the smoothed output values.

Note:

Afraid of the big, fat matrix? The method output_covariance_diagonal might just save your life.

pub fn smoothed_covariance_diagonal(&self, mix: &PPCAMix) -> DVector<f64>

Returns an approximation of the smoothed output covariance matrix, treating each masked output as an independent normal distribution.

Note:

Use this not to get lost with big matrices in the output, losing CPU, memory and hair.

pub fn extrapolated_covariance( &self, mix: &PPCAMix, sample: &MaskedSample ) -> DMatrix<f64>

The covariance for the extraplated values for a given output model and extant values in a given sample.

Note:

Afraid of the big, fat matrix? The method output_covariance_diagonal might just save your life.

pub fn extrapolated_covariance_diagonal( &self, mix: &PPCAMix, sample: &MaskedSample ) -> DVector<f64>

Returns an approximation of the extrapolated output covariance matrix, treating each masked output as an independent normal distribution.

Note

Use this not to get lost with big matrices in the output, losing CPU, memory and hair.

pub fn posterior_sampler(&self) -> PosteriorSamplerMix

Samples from the posterior distribution of an infered sample. The sample is smoothed, that is, it does not include the model isotropic noise.

Trait Implementations

impl Clone for InferredMaskedMix

fn clone(&self) -> InferredMaskedMix

Returns a copy of the value.

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

Performs copy-assignment from source.

impl Debug for InferredMaskedMix

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

Formats the value using the given formatter.

impl<'de> Deserialize<'de> for InferredMaskedMix

fn deserialize<__D>(__deserializer: __D) -> Result<Self, __D::Error>where __D: Deserializer<'de>,

Deserialize this value from the given Serde deserializer.

impl Serialize for InferredMaskedMix

fn serialize<__S>(&self, __serializer: __S) -> Result<__S::Ok, __S::Error>where __S: Serializer,

Serialize this value into the given Serde serializer.