(InflGame.domains.rd)#

resource_distribution_choice#

InflGame.domains.rd.resource_distribution_choice(bin_points, resource_type, resource_parameters)#

Select and compute a resource distribution based on the specified type and parameters.

This function serves as the main entry point for creating resource distributions in influencer games. It dispatches to the appropriate distribution function based on the resource_type parameter.

Parameters:
bin_pointsnp.ndarray

Points where the resource distribution is evaluated. For 1D distributions, this is a 1D array. For 2D distributions, this is an \((N, 2)\) array of coordinate pairs.

resource_typestr

Type of resource distribution to compute. Available options:

  • 'beta': 1D beta distribution on the interval (0,1); see beta_distribution

  • 'dirichlet_distribution': Dirichlet distribution for resources on a probability 3-simplex; see dirichlet_distribution

  • 'multi_modal_gaussian_distribution_1D': 1D mixture of Gaussian kernels; see multi_modal_gaussian_distribution_1D

  • 'multi_modal_gaussian_distribution_2D': 2D mixture of multivariate Gaussian kernels; see multi_modal_gaussian_distribution_2D

  • 'multi_modal_gaussian_distribution_2D_triangle': Alias for 2D multivariate Gaussian on triangular domain

  • 'multi_modal_gaussian_distribution_2D_square': Alias for 2D multivariate Gaussian on rectangular domain

resource_parameterslist | np.ndarray

Parameters for the specified resource distribution. Format depends on resource_type:

  • For 'beta': [alpha, beta]

  • For 'dirichlet_distribution': [alpha_1, alpha_2, ..., alpha_k]

  • For 'multi_modal_gaussian_distribution_1D': [stds, means, mode_factors]

  • For 'multi_modal_gaussian_distribution_2D': [covariance_matrices, means]

Returns:
np.ndarray

Computed resource distribution values at the specified bin_points.

Raises:
str

Returns error message string if resource_type is not recognized.

Examples

>>> bin_points = np.linspace(0, 1, 100)
>>> resources = resource_distribution_choice(bin_points, 'beta', [2, 5])
>>> resources.shape
(100,)
No-index: