Struct ndarray_glm::model::Dataset

pub struct Dataset<F> where
    F: Float, {
    pub y: Array1<F>,
    pub x: Array2<F>,
    pub linear_offset: Option<Array1<F>>,
    pub weights: Option<Array1<F>>,
    /* private fields */
}

Fields

y: Array1<F>

the observation of response data by event

x: Array2<F>

the design matrix with events in rows and instances in columns

linear_offset: Option<Array1<F>>

The offset in the linear predictor for each data point. This can be used to fix the effect of control variables.

weights: Option<Array1<F>>

The weight of each observation

Implementations

impl<F> Dataset<F> where
    F: Float, 

pub fn linear_predictor(&self, regressors: &Array1<F>) -> Array1<F>

Returns the linear predictors, i.e. the design matrix multiplied by the regression parameters. Each entry in the resulting array is the linear predictor for a given observation. If linear offsets for each observation are provided, these are added to the linear predictors

pub fn hat(&self) -> RegressionResult<Ref<'_, Array2<F>>>

Returns the hat matrix of the dataset of covariate data, also known as the “projection” or “influence” matrix.

pub fn leverage(&self) -> RegressionResult<Array1<F>>

Returns the leverage for each observation. This is given by the diagonal of the projection matrix and indicates the sensitivity of each prediction to its corresponding observation.