distributions#

Provides multiple Distributions.

class BetaDistribution(alpha: Union[float, int], beta: Union[float, int])#

Bases: object

An object which samples values from a beta distributions.

alpha#: Alpha parameter of the distributions.

beta#: Beta parameter of the distributions.

Raises:

TypeError – If alpha is not float or int.
TypeError – If beta is not float or int.
ValueError – If alpha <= 0.
ValueError – If beta <= 0.

Inits BetaDistribution with the provided arguments.

Parameters:

alpha – Alpha parameter of the distributions.
beta – Beta parameter of the distributions.

get_random_values(n: int) → list[float]#

Samples an array of values with the given shape from the distributions.

Parameters:: n – The number of values.
Returns:: an array of randomly sampled values.

get_samples_from_hypercube(n: int) → list[float]#

Samples n values from a hypercube.

Parameters:: n – the number of samples.
Returns:: Samples of the distribution, sampled from a hypercube.

class Constant(value: T)#

Bases: object

An object which represents a constant value.

value#: The value.

Raises:: TypeError – If value is not a float or int.

Inits the Constant with the provided value.

Parameters:: value – The scalar used as constant value.

get_random_values(n: int) → list[T]#

Returns an array of the constant value in the given shape.

Parameters:: n – The number of values.
Returns:: np.ndarray[float]

get_samples_from_hypercube(n: int) → list[T]#

Samples n values from a hypercube.

Parameters:: n – the number of samples.
Returns:: Samples of the distributions, sampled from a hypercube.

class ContinuousUniformDistribution(min_value: Union[float, int], max_value: Union[float, int])#

Bases: object

An object which samples values from a continuous uniform distributions.

min_value#: the minimal value of the distributions.

max_value#: the maximal value of the distributions.

Raises:

TypeError – If min_value is not float or int.
TypeError – If max_value is not float or int.

Inits ContinuousUniformDistribution with the provided arguments.

Parameters:

min_value – the minimal value of the distributions.
max_value – the maximal value of the distributions.

get_random_values(n: int) → list[float]#

Samples an array with the given distribution.

Parameters:: n – The number of values.
Returns:: np.ndarray[float]

get_samples_from_hypercube(n: int) → list[float]#

Samples n values from a hypercube.

Parameters:: n – the number of samples.
Returns:: Samples of the distributions, sampled from a hypercube.

class Distribution(*args, **kwargs)#

Bases: Protocol[T]

A Distribution presents a set of values a property can have.

Note

If no explicit way for sampling with the hypercube, the following code-snippet can probably be used: ``` exactness = 1000 vals = self.get_random_values(n * exactness) vals = np.sort(vals) return [

np.random.choice(vals[i:i+exactness ]) for i in range(0, len(vals), exactness)]

```

get_random_values(n: int) → list[T]#

Samples a random value due to the type of Distribution.

Parameters:: n – The number of values.

get_samples_from_hypercube(n: int) → list[T]#

Samples n values from a hypercube.

Parameters:: n – the number of samples.

class LogNormalDistribution(mu: Union[float, int], sigma: Union[float, int])#

Bases: object

An object which samples values from a log-normal distributions.

mu#: Mean (“centre”) of the distributions.

sigma#: Standard deviation (spread or “width”) of the distributions.

Must be non-negative.

Raises:

ValueError – If sigma < 0.
TypeError – If mu is not float or int.
TypeError – If sigma is not float or int.

Inits LogNormalDistribution with the provided arguments.

Parameters:

mu – Mean (“centre”) of the distributions.
sigma – Standard deviation (spread or “width”) of the distributions. Must be non-negative.

get_random_values(n: int) → list[float]#

Samples an array with the given distribution.

Parameters:: n – The number of values.
Returns:: np.ndarray[float]

get_samples_from_hypercube(n: int) → list[float]#

Samples n values from a hypercube.

Parameters:: n – the number of samples.
Returns:: Samples of the distribution, sampled from a hypercube.

class NormalDistribution(mu: Union[float, int], sigma: Union[float, int])#

Bases: object

An object which samples values from a normal distributions.

mu#: Mean (“centre”) of the distributions.

sigma#: Standard deviation (spread or “width”) of the distributions.

Must be non-negative.

Raises:

ValueError – If sigma < 0.
TypeError – If mu is not float or int.
TypeError – If sigma is not float or int.

Inits NormalDistribution with the provided arguments.

Parameters:

mu – Mean (“centre”) of the distributions.
sigma – Standard deviation (spread or “width”) of the distributions. Must be non-negative.

get_random_values(n: int) → list[float]#

Samples an array with the given distribution.

Parameters:: n – The number of values.
Returns:: np.ndarray[float]

get_samples_from_hypercube(n: int) → list[float]#

Samples n values from a hypercube.

Parameters:: n – the number of samples.
Returns:: Samples of the distribution, sampled from a hypercube.

class VectorDistribution(values: list[Union[float, int]])#

Bases: object

An object which samples values from a list of numbers.

values#: A list containing values.

Raises:

IndexError – If values is empty.
TypeError – If values contains a value which is not a float or int.

Inits VectorDistribution with the provided arguments.

Parameters:: values – A list containing all the valid values.

get_random_values(n: int) → list[float]#

Samples an array of values from the list of values with the given shape.

Parameters:: n – The number of values.
Returns:: np.ndarray[float]

get_samples_from_hypercube(n: int) → list[float]#

Samples n values from a hypercube.

Parameters:: n – the number of samples.
Returns:: Samples of the distributions, sampled from a hypercube.

get_random_array_from_distribution(distribution: Distribution[float], shape: tuple[int, ...]) → ndarray[Any, dtype[float64]]#

Samples a random array from the given distribution.

Parameters:

distribution – The distribution to sample from.
shape – The shape of the output array.

Returns:

A random array sampled from the distribution.

get_random_value_from_distribution(distribution: Distribution[T]) → T#

Samples a random value from the given distribution.

Parameters:: distribution – The distribution to sample from.
Returns:: A random value sampled from the distribution.

`simba_ml.simulation.distributions.beta_distribution`	Defines Beta Distribution.
`simba_ml.simulation.distributions.constant`	Defines a Distribution which only has one value.
`simba_ml.simulation.distributions.continuous_uniform_distribution`	Defines Continuous Uniform Distribution.
`simba_ml.simulation.distributions.distribution`	Defines an abstract definition of Distribution.
`simba_ml.simulation.distributions.helper_functions`	Provides functions to sample from distributions.
`simba_ml.simulation.distributions.lognormal_distribution`	Defines Lognormal Distribution.
`simba_ml.simulation.distributions.normal_distribution`	Defines the Normal Distribution.
`simba_ml.simulation.distributions.vector_distribution`	Defines Vector Distribution.