Trait ndarray_stats::QuantileExt

source · 
pub trait QuantileExt<A, S, D>where
    S: Data<Elem = A>,
    D: Dimension,{
    fn min(&self) -> Option<&A>
    where
        A: PartialOrd;
    fn min_skipnan(&self) -> &A
    where
        A: MaybeNan,
        A::NotNan: Ord;
    fn max(&self) -> Option<&A>
    where
        A: PartialOrd;
    fn max_skipnan(&self) -> &A
    where
        A: MaybeNan,
        A::NotNan: Ord;
    fn quantile_axis_mut<I>(&mut self, axis: Axis, q: f64) -> Array<A, D::Smaller>
    where
        D: RemoveAxis,
        A: Ord + Clone,
        S: DataMut,
        I: Interpolate<A>;
    fn quantile_axis_skipnan_mut<I>(
        &mut self,
        axis: Axis,
        q: f64
    ) -> Array<A, D::Smaller>
    where
        D: RemoveAxis,
        A: MaybeNan,
        A::NotNan: Clone + Ord,
        S: DataMut,
        I: Interpolate<A::NotNan>;
}
Expand description

Quantile methods for ArrayBase.

Required Methods

Finds the elementwise minimum of the array.

Returns None if any of the pairwise orderings tested by the function are undefined. (For example, this occurs if there are any floating-point NaN values in the array.)

Additionally, returns None if the array is empty.

Finds the elementwise minimum of the array, skipping NaN values.

Warning This method will return a NaN value if none of the values in the array are non-NaN values. Note that the NaN value might not be in the array.

Finds the elementwise maximum of the array.

Returns None if any of the pairwise orderings tested by the function are undefined. (For example, this occurs if there are any floating-point NaN values in the array.)

Additionally, returns None if the array is empty.

Finds the elementwise maximum of the array, skipping NaN values.

Warning This method will return a NaN value if none of the values in the array are non-NaN values. Note that the NaN value might not be in the array.

Return the qth quantile of the data along the specified axis.

q needs to be a float between 0 and 1, bounds included. The qth quantile for a 1-dimensional lane of length N is defined as the element that would be indexed as (N-1)q if the lane were to be sorted in increasing order. If (N-1)q is not an integer the desired quantile lies between two data points: we return the lower, nearest, higher or interpolated value depending on the type Interpolate bound I.

Some examples:

  • q=0. returns the minimum along each 1-dimensional lane;
  • q=0.5 returns the median along each 1-dimensional lane;
  • q=1. returns the maximum along each 1-dimensional lane. (q=0 and q=1 are considered improper quantiles)

The array is shuffled in place along each 1-dimensional lane in order to produce the required quantile without allocating a copy of the original array. Each 1-dimensional lane is shuffled independently from the others. No assumptions should be made on the ordering of the array elements after this computation.

Complexity (quickselect):

  • average case: O(m);
  • worst case: O(m^2); where m is the number of elements in the array.

Panics if axis is out of bounds, if the axis has length 0, or if q is not between 0. and 1. (inclusive).

Return the qth quantile of the data along the specified axis, skipping NaN values.

See quantile_axis_mut for details.

Implementations on Foreign Types

Implementors