Struct ndarray_stats::histogram::Grid
source · [−]pub struct Grid<A: Ord> { /* private fields */ }
Expand description
An orthogonal partition of a rectangular region in an n-dimensional space, e.g. [a0, b0) × ⋯ × [an−1, bn−1), represented as a collection of rectangular n-dimensional bins.
The grid is solely determined by the Cartesian product of its projections on each coordinate axis. Therefore, each element in the product set should correspond to a sub-region in the grid.
For example, this partition can be represented as a Grid
struct:
g +---+-------+---+
| 3 | 4 | 5 |
f +---+-------+---+
| | | |
| 0 | 1 | 2 |
| | | |
e +---+-------+---+
a b c d
R0: [a, b) × [e, f)
R1: [b, c) × [e, f)
R2: [c, d) × [e, f)
R3: [a, b) × [f, g)
R4: [b, d) × [f, g)
R5: [c, d) × [f, g)
Grid: { [a, b), [b, c), [c, d) } × { [e, f), [f, g) } == { R0, R1, R2, R3, R4, R5 }
while the next one can’t:
g +---+-----+---+
| | 2 | 3 |
(f) | +-----+---+
| 0 | |
| | 1 |
| | |
e +---+-----+---+
a b c d
R0: [a, b) × [e, g)
R1: [b, d) × [e, f)
R2: [b, c) × [f, g)
R3: [c, d) × [f, g)
// 'f', as long as 'R1', 'R2', or 'R3', doesn't appear on LHS
// [b, c) × [e, g), [c, d) × [e, g) doesn't appear on RHS
Grid: { [a, b), [b, c), [c, d) } × { [e, g) } != { R0, R1, R2, R3 }
Examples
Basic usage, building a Grid
via GridBuilder
, with optimal grid layout determined by
a given strategy
, and generating a histogram
:
use ndarray::{Array, array};
use ndarray_stats::{
histogram::{strategies::Auto, Bins, Edges, Grid, GridBuilder},
HistogramExt,
};
// 1-dimensional observations, as a (n_observations, n_dimension) 2-d matrix
let observations = Array::from_shape_vec(
(12, 1),
vec![1, 4, 5, 2, 100, 20, 50, 65, 27, 40, 45, 23],
).unwrap();
// The optimal grid layout is inferred from the data, given a chosen strategy, Auto in this case
let grid = GridBuilder::<Auto<usize>>::from_array(&observations).unwrap().build();
let histogram = observations.histogram(grid);
let histogram_matrix = histogram.counts();
// Bins are left-closed, right-open!
let expected = array![4, 3, 3, 1, 0, 1];
assert_eq!(histogram_matrix, expected.into_dyn());
Implementations
sourceimpl<A: Ord> Grid<A>
impl<A: Ord> Grid<A>
sourcepub fn ndim(&self) -> usize
pub fn ndim(&self) -> usize
Returns the number of dimensions of the region partitioned by the grid.
Examples
use ndarray_stats::histogram::{Edges, Bins, Grid};
let edges = Edges::from(vec![0, 1]);
let bins = Bins::new(edges);
let square_grid = Grid::from(vec![bins.clone(), bins.clone()]);
assert_eq!(square_grid.ndim(), 2usize)
sourcepub fn shape(&self) -> Vec<usize>
pub fn shape(&self) -> Vec<usize>
Returns the numbers of bins along each coordinate axis.
Examples
use ndarray_stats::histogram::{Edges, Bins, Grid};
let edges_x = Edges::from(vec![0, 1]);
let edges_y = Edges::from(vec![-1, 0, 1]);
let bins_x = Bins::new(edges_x);
let bins_y = Bins::new(edges_y);
let square_grid = Grid::from(vec![bins_x, bins_y]);
assert_eq!(square_grid.shape(), vec![1usize, 2usize]);
sourcepub fn projections(&self) -> &[Bins<A>]
pub fn projections(&self) -> &[Bins<A>]
Returns the grid projections on each coordinate axis as a slice of immutable references.
sourcepub fn index_of<S>(&self, point: &ArrayBase<S, Ix1>) -> Option<Vec<usize>> where
S: Data<Elem = A>,
pub fn index_of<S>(&self, point: &ArrayBase<S, Ix1>) -> Option<Vec<usize>> where
S: Data<Elem = A>,
Returns an n-dimensional
index, of bins along each axis that contains the point, if one
exists.
Returns None
if the point is outside the grid.
Panics
Panics if dimensionality of the point doesn’t equal the grid’s.
Examples
Basic usage:
use ndarray::array;
use ndarray_stats::histogram::{Edges, Bins, Grid};
use noisy_float::types::n64;
let edges = Edges::from(vec![n64(-1.), n64(0.), n64(1.)]);
let bins = Bins::new(edges);
let square_grid = Grid::from(vec![bins.clone(), bins.clone()]);
// (0., -0.7) falls in 1st and 0th bin respectively
assert_eq!(
square_grid.index_of(&array![n64(0.), n64(-0.7)]),
Some(vec![1, 0]),
);
// Returns `None`, as `1.` is outside the grid since bins are right-open
assert_eq!(
square_grid.index_of(&array![n64(0.), n64(1.)]),
None,
);
A panic upon dimensionality mismatch:
// the point has 3 dimensions, the grid expected 2 dimensions
assert_eq!(
square_grid.index_of(&array![n64(0.), n64(-0.7), n64(0.5)]),
Some(vec![1, 0, 1]),
);
sourceimpl<A: Ord + Clone> Grid<A>
impl<A: Ord + Clone> Grid<A>
sourcepub fn index(&self, index: &[usize]) -> Vec<Range<A>>
pub fn index(&self, index: &[usize]) -> Vec<Range<A>>
Given an n
-dimensional index, i = (i_0, ..., i_{n-1})
, returns an n
-dimensional bin,
I_{i_0} x ... x I_{i_{n-1}}
, where I_{i_j}
is the i_j
-th interval on the j
-th
projection of the grid on the coordinate axes.
Panics
Panics if at least one in the index, (i_0, ..., i_{n-1})
, is out of bounds on the
corresponding coordinate axis, i.e. if there exists j
s.t.
i_j >= self.projections[j].len()
.
Examples
Basic usage:
use ndarray::array;
use ndarray_stats::histogram::{Edges, Bins, Grid};
let edges_x = Edges::from(vec![0, 1]);
let edges_y = Edges::from(vec![2, 3, 4]);
let bins_x = Bins::new(edges_x);
let bins_y = Bins::new(edges_y);
let square_grid = Grid::from(vec![bins_x, bins_y]);
// Query the 0-th bin on x-axis, and 1-st bin on y-axis
assert_eq!(
square_grid.index(&[0, 1]),
vec![0..1, 3..4],
);
A panic upon out-of-bounds:
// out-of-bound on y-axis
assert_eq!(
square_grid.index(&[0, 2]),
vec![0..1, 3..4],
);
Trait Implementations
sourceimpl<A: Ord> From<Vec<Bins<A>, Global>> for Grid<A>
impl<A: Ord> From<Vec<Bins<A>, Global>> for Grid<A>
sourcefn from(projections: Vec<Bins<A>>) -> Self
fn from(projections: Vec<Bins<A>>) -> Self
Converts a Vec<Bins<A>>
into a Grid<A>
, consuming the vector of bins.
The i
-th element in Vec<Bins<A>>
represents the projection of the bin grid onto the
i
-th axis.
Alternatively, a Grid
can be built directly from data using a GridBuilder
.
sourceimpl<A: PartialEq + Ord> PartialEq<Grid<A>> for Grid<A>
impl<A: PartialEq + Ord> PartialEq<Grid<A>> for Grid<A>
impl<A: Eq + Ord> Eq for Grid<A>
impl<A: Ord> StructuralEq for Grid<A>
impl<A: Ord> StructuralPartialEq for Grid<A>
Auto Trait Implementations
impl<A> RefUnwindSafe for Grid<A> where
A: RefUnwindSafe,
impl<A> Send for Grid<A> where
A: Send,
impl<A> Sync for Grid<A> where
A: Sync,
impl<A> Unpin for Grid<A> where
A: Unpin,
impl<A> UnwindSafe for Grid<A> where
A: UnwindSafe,
Blanket Implementations
sourceimpl<T> BorrowMut<T> for T where
T: ?Sized,
impl<T> BorrowMut<T> for T where
T: ?Sized,
const: unstable · sourcefn borrow_mut(&mut self) -> &mut T
fn borrow_mut(&mut self) -> &mut T
Mutably borrows from an owned value. Read more
sourceimpl<Q, K> Equivalent<K> for Q where
Q: Eq + ?Sized,
K: Borrow<Q> + ?Sized,
impl<Q, K> Equivalent<K> for Q where
Q: Eq + ?Sized,
K: Borrow<Q> + ?Sized,
sourcefn equivalent(&self, key: &K) -> bool
fn equivalent(&self, key: &K) -> bool
Compare self to key
and return true
if they are equal.