projnormal.formulas.projected_normal_c.sampling

Sampling functions for the general projected normal distribution with an additive constant const in the denominator.

Functions

empirical_moments(mean_x, covariance_x, ...)

Compute the mean, covariance and second moment of the variable \(y = x/\sqrt{x^T x + c}\) where \(x \sim \mathcal{N}(\mu_x, \Sigma_x)\) and \(c\) is a positive constant added to the denominator, by sampling from the distribution.

sample(mean_x, covariance_x, const, n_samples)

Sample the variable \(y = x/\sqrt{x^T x + c}\) where \(x \sim \mathcal{N}(\mu_x, \Sigma_x)\) and \(c\) is a constant added to the denominator.

empirical_moments(mean_x, covariance_x, const, n_samples)

Compute the mean, covariance and second moment of the variable \(y = x/\sqrt{x^T x + c}\) where \(x \sim \mathcal{N}(\mu_x, \Sigma_x)\) and \(c\) is a positive constant added to the denominator, by sampling from the distribution.

Parameters:
  • mean_x (torch.Tensor) – Mean of x. Shape is (n_dim,).

  • covariance_x (torch.Tensor) – Covariance of x. Shape is (n_dim, n_dim).

  • const (torch.Tensor) – Constant added to the denominator. Shape is ().

  • n_samples (int) – Number of samples to draw.

Returns:

Dictionary with the keys mean, covariance, and second_moment, containing the empirical moments of the projected normal distribution.

Return type:

dict

sample(mean_x, covariance_x, const, n_samples)

Sample the variable \(y = x/\sqrt{x^T x + c}\) where \(x \sim \mathcal{N}(\mu_x, \Sigma_x)\) and \(c\) is a constant added to the denominator.

Parameters:
  • mean_x (torch.Tensor) – Mean of x. Shape is (n_dim,).

  • covariance_x (torch.Tensor) – Covariance of x. Shape is (n_dim, n_dim).

  • const (torch.Tensor) – Constant added to the denominator. Shape is ().

  • n_samples (int) – Number of samples to draw.

Returns:

Samples from the distribution. Shape is (n_samples, n_dim).

Return type:

torch.Tensor