Table of Contents

Class PolynomialModule

Namespace
NumSharp
Assembly
NumSharp.dll

The numpy.polynomial package facade, reachable as polynomial. Its members are the six basis submodules; each is a first-class object, so Python's import … as P becomes var P = np.polynomial.polynomial;. (A nested static class cannot model this: C# forbids a member named polynomial inside a type of that name.)

[ModuleName("np.polynomial")]
public class PolynomialModule
Inheritance
PolynomialModule
Inherited Members

Remarks

Implemented so far, bit-identical to NumPy 2.4.2 for every dtype NumPy has (values, dtypes, result layouts and error texts; the one dtype divergence is NumSharp's single complex width, #569):

  • the evaluation family of every basis — {p}val, {p}val2d, {p}val3d, {p}grid2d, {p}grid3d and {p}valnd — through one runtime-compiled IL kernel per (basis, dtypes, layout) class, the result laid out as NumPy's last ufunc call lays it out (an F-ordered or permuted x or series keeps its memory order);
  • the additive family of every basis — the constants {p}domain/{p}zero/{p}one/ {p}x, {p}line, {p}add, {p}sub, {p}trim — and the polyutils substrate (as_series, trimseq, trimcoef, getdomain, mapparms, mapdomain);
  • the calculus family of every basis — {p}der and {p}int with every parameter (m, k, lbnd, scl, axis) — through one runtime-compiled IL kernel per (basis, direction, dtypes) that runs every order in place in one buffer;
  • the series algebra of every basis — {p}mulx, {p}mul, {p}div, {p}pow, {p}fromroots and the X2poly/poly2X conversions — NumPy's Python statements replayed over a pooled per-thread arena, each statement on the house kernel NumPy's statement runs through (long products byte-exact with NumSharp.Interop.OpenBLAS, bounded-ULP without it);
  • the Vandermonde family of every basis — {p}vander, {p}vander2d, {p}vander3d — one whole-array kernel per (basis, dtype, rank), returning NumPy's moveaxis view;
  • companion matrices and roots of every basis — {p}companion, {p}roots (roots of degree ≥ 2 run np.linalg.eigvals, so they need a LAPACK backend such as NumSharp.Interop.OpenBLAS);
  • the weight functions {p}weight, the Chebyshev points and Gauss quadratures (chebpts1, chebpts2, chebgauss, leggauss, laggauss, hermgauss, hermegauss; the last four run np.linalg.eigvalsh), chebinterpolate, the least-squares fits {p}fit (np.linalg.lstsq) and polyvalfromroots;
  • printing: set_default_printstyle(object) and format_float(object, bool);
  • the six series classes — NDPolynomial, NDChebyshev, NDLegendre, NDLaguerre, NDHermite, NDHermiteE (factories Polynomial(object, object, object, string), Chebyshev(object, object, object, string), …) — NumPy's ABCPolyBase statement for statement over the module functions: the arithmetic with Python's operator protocol, evaluation (at a NumPy scalar in scalarmath, as NumPy's p(x) does), calculus, roots, fits, basis conversion (NumPy's recurrence run at a series) and printing.

The kernels are emitted at run time and have no fallback: under NativeAOT (no dynamic code) a call that reaches one raises PlatformNotSupportedException, and the evaluation, calculus, algebra, Vandermonde and companion paths all reach one.

https://numpy.org/doc/stable/reference/routines.polynomials-package.html

Properties

chebyshev

numpy.polynomial.chebyshev — Chebyshev series of the first kind.

public ChebyshevModule chebyshev { get; }

Property Value

ChebyshevModule

hermite

numpy.polynomial.hermite — physicists' Hermite series.

public HermiteModule hermite { get; }

Property Value

HermiteModule

hermite_e

numpy.polynomial.hermite_e — probabilists' Hermite series.

public HermiteEModule hermite_e { get; }

Property Value

HermiteEModule

laguerre

numpy.polynomial.laguerre — Laguerre series.

public LaguerreModule laguerre { get; }

Property Value

LaguerreModule

legendre

numpy.polynomial.legendre — Legendre series.

public LegendreModule legendre { get; }

Property Value

LegendreModule

polynomial

numpy.polynomial.polynomial — power series c[0] + c[1]*x + c[2]*x**2 + … (coefficients low degree first; see PowerSeriesModule).

public PowerSeriesModule polynomial { get; }

Property Value

PowerSeriesModule

polyutils

numpy.polynomial.polyutils — the helpers every basis is built on (series normalization and the linear domain maps; see PolyUtilsModule).

public PolyUtilsModule polyutils { get; }

Property Value

PolyUtilsModule

Methods

Chebyshev(object, object, object, string)

NumPy's numpy.polynomial.chebyshev.Chebyshev(coef, domain=None, window=None, symbol='x'): a Chebyshev series — a new NDChebyshev (the class itself; its class methods fit, fromroots, identity, basis and cast are static members of NDChebyshev).

public NDChebyshev Chebyshev(object coef, object domain = null, object window = null, string symbol = "x")

Parameters

coef object

The coefficients, low degree first (see NDChebyshev(object, object, object, string)).

domain object

The domain, or null for the class default.

window object

The window, or null for the class default.

symbol string

The variable's name for printing: a Python identifier.

Returns

NDChebyshev

The series (dispose it when done).

Remarks

Exceptions

ValueError

The constructor's errors (an empty or non-1-d series, a domain or window of the wrong length, an invalid symbol).

TypeError

A null symbol.

NotSupportedException

Coefficients NumPy keeps as an object array.

Hermite(object, object, object, string)

NumPy's numpy.polynomial.hermite.Hermite(coef, domain=None, window=None, symbol='x'): a (physicists') Hermite series — a new NDHermite (the class itself; its class methods fit, fromroots, identity, basis and cast are static members of NDHermite).

public NDHermite Hermite(object coef, object domain = null, object window = null, string symbol = "x")

Parameters

coef object

The coefficients, low degree first (see NDHermite(object, object, object, string)).

domain object

The domain, or null for the class default.

window object

The window, or null for the class default.

symbol string

The variable's name for printing: a Python identifier.

Returns

NDHermite

The series (dispose it when done).

Remarks

Exceptions

ValueError

The constructor's errors (an empty or non-1-d series, a domain or window of the wrong length, an invalid symbol).

TypeError

A null symbol.

NotSupportedException

Coefficients NumPy keeps as an object array.

HermiteE(object, object, object, string)

NumPy's numpy.polynomial.hermite_e.HermiteE(coef, domain=None, window=None, symbol='x'): a (probabilists') HermiteE series — a new NDHermiteE (the class itself; its class methods fit, fromroots, identity, basis and cast are static members of NDHermiteE).

public NDHermiteE HermiteE(object coef, object domain = null, object window = null, string symbol = "x")

Parameters

coef object

The coefficients, low degree first (see NDHermiteE(object, object, object, string)).

domain object

The domain, or null for the class default.

window object

The window, or null for the class default.

symbol string

The variable's name for printing: a Python identifier.

Returns

NDHermiteE

The series (dispose it when done).

Remarks

Exceptions

ValueError

The constructor's errors (an empty or non-1-d series, a domain or window of the wrong length, an invalid symbol).

TypeError

A null symbol.

NotSupportedException

Coefficients NumPy keeps as an object array.

Laguerre(object, object, object, string)

NumPy's numpy.polynomial.laguerre.Laguerre(coef, domain=None, window=None, symbol='x'): a Laguerre series — a new NDLaguerre (the class itself; its class methods fit, fromroots, identity, basis and cast are static members of NDLaguerre).

public NDLaguerre Laguerre(object coef, object domain = null, object window = null, string symbol = "x")

Parameters

coef object

The coefficients, low degree first (see NDLaguerre(object, object, object, string)).

domain object

The domain, or null for the class default.

window object

The window, or null for the class default.

symbol string

The variable's name for printing: a Python identifier.

Returns

NDLaguerre

The series (dispose it when done).

Remarks

Exceptions

ValueError

The constructor's errors (an empty or non-1-d series, a domain or window of the wrong length, an invalid symbol).

TypeError

A null symbol.

NotSupportedException

Coefficients NumPy keeps as an object array.

Legendre(object, object, object, string)

NumPy's numpy.polynomial.legendre.Legendre(coef, domain=None, window=None, symbol='x'): a Legendre series — a new NDLegendre (the class itself; its class methods fit, fromroots, identity, basis and cast are static members of NDLegendre).

public NDLegendre Legendre(object coef, object domain = null, object window = null, string symbol = "x")

Parameters

coef object

The coefficients, low degree first (see NDLegendre(object, object, object, string)).

domain object

The domain, or null for the class default.

window object

The window, or null for the class default.

symbol string

The variable's name for printing: a Python identifier.

Returns

NDLegendre

The series (dispose it when done).

Remarks

Exceptions

ValueError

The constructor's errors (an empty or non-1-d series, a domain or window of the wrong length, an invalid symbol).

TypeError

A null symbol.

NotSupportedException

Coefficients NumPy keeps as an object array.

Polynomial(object, object, object, string)

NumPy's numpy.polynomial.polynomial.Polynomial(coef, domain=None, window=None, symbol='x'): a power series — a new NDPolynomial (the class itself; its class methods fit, fromroots, identity, basis and cast are static members of NDPolynomial).

public NDPolynomial Polynomial(object coef, object domain = null, object window = null, string symbol = "x")

Parameters

coef object

The coefficients, low degree first (see NDPolynomial(object, object, object, string)).

domain object

The domain, or null for the class default.

window object

The window, or null for the class default.

symbol string

The variable's name for printing: a Python identifier.

Returns

NDPolynomial

The series (dispose it when done).

Remarks

Exceptions

ValueError

The constructor's errors (an empty or non-1-d series, a domain or window of the wrong length, an invalid symbol).

TypeError

A null symbol.

NotSupportedException

Coefficients NumPy keeps as an object array.

set_default_printstyle(object)

Sets how the series classes print their terms (NumPy's numpy.polynomial.set_default_printstyle): "unicode" writes superscripts and subscripts (1.0 + 2.0·x + 3.0·x², T₁(x)), "ascii" plain text (1.0 + 2.0 x + 3.0 x**2, T_1(x)).

public void set_default_printstyle(object style)

Parameters

style object

"unicode" or "ascii", case-sensitive; a one-element U / StringDType array holding one of them is read as its element, as NumPy's == reads it.

Remarks

The style is ONE process-wide setting, like NumPy's class attribute ABCPolyBase._use_unicode (not scoped to a thread). NumPy's default is os.name != 'nt': ASCII on Windows, Unicode elsewhere — NumSharp starts from the same platform rule.

The series classes read it in str(p) (__str__(), ToString()); format(p, "ascii") / "unicode" (__format__(string)) choose per call.

https://numpy.org/doc/stable/reference/generated/numpy.polynomial.set_default_printstyle.html

Exceptions

ValueError

Any other style: Unsupported format string '{str(style)}'. Valid options are 'ascii' and 'unicode' — also for a number, null ('None'), bytes or a list; an empty array or one of two or more elements raises NumPy's truth-value ValueError instead.