Struct BinnedGuideTerm

pub struct BinnedGuideTerm { /* private fields */ }

A LikelihoodTerm whose size is proportional to the χ²-distance from a binned projection of the fit to a provided set of datapoints representing the true values in each bin.

This is intended to be used as follows. Suppose we perform a binned fit to a simple amplitude which is not parameterized over the binning variable. We then form a new Model which is parameterized over said variable, and we wish to perform an unbinned fit. If we can isolate terms which are not interfering, we could imagine fitting the unbinned data with a cost function that minimizes the distance to the result from the binned fit. From there, it is up to the user to decide what to do with this minimum. Caution should be used, as this will not be the minimum of the NLL, but of the guide term only. However, this minimum could be used as an intermediate for getting close to a global minimum if the likelihood landscape has many local minima. Then a true fit could be performed, starting at this intermediate point.

Implementations

impl BinnedGuideTerm

pub fn new<V, L, T, U, E>( nll: Box<NLL>, variable: &V, amplitude_sets: &[T], bins: usize, range: (f64, f64), count_sets: &[U], error_sets: Option<&[E]>, ) -> Box<BinnedGuideTerm>

where V: Variable + 'static, L: AsRef<str>, T: AsRef<[L]>, U: AsRef<[f64]>, E: AsRef<[f64]>,

Construct a new BinnedGuideTerm

This term takes a list of subsets of amplitudes, activates each set, and compares the projected histogram to the known one provided at construction. Both count_sets and error_sets should have the same shape, and their first dimension should be the same as that of amplitude_sets.

The intended usage is to provide some sets of amplitudes to isolate, like [["amp1", "amp2"], ["amp3"]], along with some known counts for a binned fit (count_sets ~ [[histogram counts involving "amp1" and "amp2"], [histogram counts involving "amp3"]] and simlar for error_sets).

Trait Implementations

impl Clone for BinnedGuideTerm

fn clone(&self) -> BinnedGuideTerm

Returns a duplicate of the value.

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

Performs copy-assignment from source.

impl LikelihoodTerm for BinnedGuideTerm

fn evaluate(&self, parameters: &[f64]) -> f64

Evaluate the term given some input parameters.

fn parameters(&self) -> Vec<String>

The list of names of the input parameters for LikelihoodTerm::evaluate.

fn evaluate_gradient( &self, parameters: &[f64], ) -> Matrix<f64, Dyn, Const<1>, VecStorage<f64, Dyn, Const<1>>>

Evaluate the gradient of the term given some input parameters.