regressioninc.testing.complex module#

Functions for generating and visualising complex-valued testing data

pydantic model regressioninc.testing.complex.ComplexGrid[source]#

Bases: BaseModel

A class to generate a grid for complex data

Show JSON schema

{
   "title": "ComplexGrid",
   "description": "A class to generate a grid for complex data",
   "type": "object",
   "properties": {
      "r1": {
         "title": "R1",
         "type": "number"
      },
      "r2": {
         "title": "R2",
         "type": "number"
      },
      "nr": {
         "title": "Nr",
         "type": "integer"
      },
      "i1": {
         "title": "I1",
         "type": "number"
      },
      "i2": {
         "title": "I2",
         "type": "number"
      },
      "ni": {
         "title": "Ni",
         "type": "integer"
      }
   },
   "required": [
      "r1",
      "r2",
      "nr",
      "i1",
      "i2",
      "ni"
   ]
}

field r1: float [Required]#: Starting real pt

field r2: float [Required]#: End real pt

field nr: int [Required]#: Number of real pts

field i1: float [Required]#: Start imaginary pt

field i2: float [Required]#: End imaginary pt

field ni: int [Required]#: Number of imaginary pts

property n_pts: int#

Get the number of points grid

Returns:: Number of points in grid
Return type:: int

Examples

>>> from regressioninc.testing.complex import ComplexGrid
>>> grid = ComplexGrid(r1=0, r2=5, nr=6, i1=0, i2=5, ni=6)
>>> grid.n_pts
36

grid() → ndarray[source]#

Get the grid points as an 2-D array

Returns:: The grid points as a 2-D array
Return type:: np.ndarray

Examples

>>> from regressioninc.testing.complex import ComplexGrid
>>> grid = ComplexGrid(r1=-1, r2=1, nr=3, i1=-1, i2=1, ni=3)
>>> grid.grid()
array([[-1.-1.j, -1.+0.j, -1.+1.j],
       [ 0.-1.j,  0.+0.j,  0.+1.j],
       [ 1.-1.j,  1.+0.j,  1.+1.j]])

flat_grid() → ndarray[source]#

Get the grid as a flat array

Returns:: The grid of points as a 1-D flattened array
Return type:: np.ndarray

Examples

>>> from regressioninc.testing.complex import ComplexGrid
>>> grid = ComplexGrid(r1=-1, r2=1, nr=3, i1=-1, i2=1, ni=3)
>>> grid.grid()
array([[-1.-1.j, -1.+0.j, -1.+1.j],
       [ 0.-1.j,  0.+0.j,  0.+1.j],
       [ 1.-1.j,  1.+0.j,  1.+1.j]])

regressioninc.testing.complex.generate_linear_grid(coef: ndarray, grids: list[regressioninc.testing.complex.ComplexGrid], intercept: complex = 0) → tuple[numpy.ndarray, numpy.ndarray][source]#

Generate complex regression data from coefficients and grids of regressors

Parameters:

coef (np.ndarray) – The coefficients for the regressors
grids (list[ComplexGrid]) – The grid of points for each regressor
intercept (complex, optional) – The intercept, by default 0

Returns:

X, y for the regression problem

Return type:

tuple[np.ndarray, np.ndarray]

Raises:

ValueError – If the number of grids provided does not equal the number of coefficients
ValueError – If the grids have different numbers of points

regressioninc.testing.complex.generate_linear_random(coef: ndarray, n_samples: int, intercept: complex = 0, min_rand=-10, max_rand=10)[source]#: Produce complex data for testing without any noise

regressioninc.testing.complex.add_gaussian_noise(data: ndarray, loc: tuple[float] | None = None, scale: tuple[float] | None = None) → ndarray[source]#: Add Gaussian noise to data

regressioninc.testing.complex.add_outliers(y: ndarray, outlier_percent: float = 5, mult_min=3, mult_max=5, random_signs_real: bool = False, random_signs_imag: bool = False) → ndarray[source]#: Add outliers to a complex-valued 1-D observations array

regressioninc.testing.complex.plot_observations(y: ndarray, size: int = 10, alpha: float = 1.0) → None[source]#: Plot observation data

regressioninc.testing.complex.plot_observations_original(y_orig: ndarray, size: int = 10, alpha: float = 1.0) → None[source]#: Plot original observations, meant for data without noise

regressioninc.testing.complex.plot_regressor(reg: ndarray, color: ndarray, size: int = 10, alpha: float = 1.0) → None[source]#: Plot regression data

regressioninc.testing.complex.plot_estimate(est: ndarray, color: ndarray, size: int = 10, alpha: float = 1.0, label: str = 'estimate') → None[source]#: Plot model estimates

regressioninc.testing.complex.plot_complex(X, y, models: dict[str, regressioninc.base.Regressor], y_orig: ndarray | None = None, size_obs: int = 10, size_reg: int = 10, size_est: int = 10)[source]#: Plot the complex data