Struct AuxRichnessMetrics 

pub struct AuxRichnessMetrics {
    pub aux_observed: bool,
    pub n_nonfinite_aux: usize,
    pub aux_dim: usize,
    pub latent_dim: usize,
    pub n_rows: usize,
    pub constant_columns: Vec<usize>,
    pub aux_is_discrete: bool,
    pub n_distinct_levels: usize,
    pub jacobian_rank: usize,
    pub jacobian_rank_estimated: bool,
}

Scalar facts about the auxiliary covariate / latent pair feeding an iVAE.

Fields

aux_observed: bool

true iff every entry of the aux matrix is finite.

n_nonfinite_aux: usize

Number of non-finite entries in the aux matrix.

aux_dim: usize

Aux dimension (column count).

latent_dim: usize

Latent dimension (column count of latents).

n_rows: usize

Row count N.

constant_columns: Vec<usize>

Column indices (sorted, ascending) that are constant across rows.

aux_is_discrete: bool

true iff aux is integer-valued and every column has <= 64 unique values.

n_distinct_levels: usize

Joint distinct-row count of aux (only computed when aux_is_discrete).

jacobian_rank: usize

Empirical rank of the least-squares Jacobian B = (Aᵀ A)^{-1} Aᵀ Z. usize::MAX sentinel if the rank could not be estimated (e.g. too few rows).

jacobian_rank_estimated: bool

True iff we had enough rows + finite data to estimate the Jacobian rank.

Trait Implementations

impl Clone for AuxRichnessMetrics

fn clone(&self) -> AuxRichnessMetrics

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for AuxRichnessMetrics

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

Formats the value using the given formatter.