Struct Design 

pub struct Design { /* private fields */ }

A design matrix (rows are observations, columns are coefficients), with an offset and prior weights.

Stored column by column. The offset (default 0) enters the linear predictor with a fixed coefficient of 1, as ln(exposure) does in a log-link frequency model; the weights (default 1) divide the variance.

Implementations

impl Design

pub fn new(names: Vec<String>, columns: Vec<Vec<f64>>) -> Result<Self>

A design from named columns. Fails if they differ in length, are empty, or hold a non-finite value.

pub fn with_offset(self, offset: Vec<f64>) -> Result<Self>

Sets the offset, one value per row.

pub fn with_weights(self, weights: Vec<f64>) -> Result<Self>

Sets the prior weights, one positive value per row.

pub fn names(&self) -> &[String]

Column names.

pub fn n_rows(&self) -> usize

Number of rows.

pub fn n_cols(&self) -> usize

Number of columns.

pub fn column(&self, j: usize) -> &[f64]

Column j.

pub fn offset(&self) -> &[f64]

Offset per row.

pub fn weights(&self) -> &[f64]

Prior weight per row.

pub fn linear_predictor(&self, beta: &[f64]) -> Vec<f64>

X β + offset.

pub fn select(&self, rows: &[usize]) -> Self

The rows in rows, in that order, with their offsets and weights.

Trait Implementations

impl Clone for Design

fn clone(&self) -> Self

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl Debug for Design

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

Formats the value using the given formatter.

impl PartialEq for Design

fn eq(&self, other: &Self) -> bool

Equality operator ==.

fn ne(&self, other: &Rhs) -> bool

Inequality operator !=.

impl StructuralPartialEq for Design