pub trait Determination: Predictive {
    // Provided methods
    fn determination(
        &self,
        predictors: impl Iterator<Item = f64>,
        outcomes: impl Iterator<Item = f64> + Clone,
        len: usize
    ) -> f64 { ... }
    fn determination_slice(&self, predictors: &[f64], outcomes: &[f64]) -> f64 { ... }
}
Available on crate feature regression only.
Expand description

Helper trait to make the method take a generic iterator.

This enables Predictive to be dyn.

Provided Methods§

source

fn determination( &self, predictors: impl Iterator<Item = f64>, outcomes: impl Iterator<Item = f64> + Clone, len: usize ) -> f64

Calculates the R² (coefficient of determination), the proportion of variation in predicted model.

predictors are the x values (input to the function). outcomes are the observed dependant variable. len is the count of data points.

If predictors and outcomes have different lengths, the result might be unexpected.

O(n)

source

fn determination_slice(&self, predictors: &[f64], outcomes: &[f64]) -> f64

Convenience method for Determination::determination when using slices.

Implementors§