colosseum.noises.student_t_correlated 

 1from typing import Callable, List
 2
 3import numpy as np
 4from scipy.stats import wishart, multivariate_t
 5
 6from colosseum.noises.base import Noise
 7
 8
 9class StudentTCorrelated(Noise):
10    """
11    The class that creates Student's t correlated noise.
12    """
13
14    def _sample_noise(self, n: int) -> np.ndarray:
15        if self.rv is None:
16            W = wishart(scale=[self._scale] * np.prod(self.shape)).rvs(1, self._rng)
17            self.rv = multivariate_t(shape=W)
18        return self.rv.rvs(n, self._rng).reshape(n, *self.shape)
19
20    def __init__(self, seed: int, shape_f: Callable[[], List[int]], scale: float = 0.1):
21        """
22        Parameters
23        ----------
24        seed : int
25            The random seed.
26        shape_f : Callable[[], List[int]]
27            The function that returns the shape of the emission map.
28        scale : float
29            The scale parameter for the Wishart distribution for the covariance matrix. By default, it is 0.1.
30        """
31        super(StudentTCorrelated, self).__init__(seed, shape_f)
32
33        self._scale = scale
34        self.rv = None
class StudentTCorrelated(colosseum.noises.base.Noise): 
10class StudentTCorrelated(Noise):
11    """
12    The class that creates Student's t correlated noise.
13    """
14
15    def _sample_noise(self, n: int) -> np.ndarray:
16        if self.rv is None:
17            W = wishart(scale=[self._scale] * np.prod(self.shape)).rvs(1, self._rng)
18            self.rv = multivariate_t(shape=W)
19        return self.rv.rvs(n, self._rng).reshape(n, *self.shape)
20
21    def __init__(self, seed: int, shape_f: Callable[[], List[int]], scale: float = 0.1):
22        """
23        Parameters
24        ----------
25        seed : int
26            The random seed.
27        shape_f : Callable[[], List[int]]
28            The function that returns the shape of the emission map.
29        scale : float
30            The scale parameter for the Wishart distribution for the covariance matrix. By default, it is 0.1.
31        """
32        super(StudentTCorrelated, self).__init__(seed, shape_f)
33
34        self._scale = scale
35        self.rv = None

The class that creates Student's t correlated noise.

StudentTCorrelated(seed: int, shape_f: Callable[[], List[int]], scale: float = 0.1) 
21    def __init__(self, seed: int, shape_f: Callable[[], List[int]], scale: float = 0.1):
22        """
23        Parameters
24        ----------
25        seed : int
26            The random seed.
27        shape_f : Callable[[], List[int]]
28            The function that returns the shape of the emission map.
29        scale : float
30            The scale parameter for the Wishart distribution for the covariance matrix. By default, it is 0.1.
31        """
32        super(StudentTCorrelated, self).__init__(seed, shape_f)
33
34        self._scale = scale
35        self.rv = None
Parameters
  • seed (int): The random seed.
  • shape_f (Callable[[], List[int]]): The function that returns the shape of the emission map.
  • scale (float): The scale parameter for the Wishart distribution for the covariance matrix. By default, it is 0.1.
Inherited Members
colosseum.noises.base.Noise
shape