Nearest Neighbors

kerch.method.knn(dists: Tensor, observations: Tensor, num: int = 1) → Tensor[source]

For each distance dists, returns the average of the num smallest corresponding observations.

Parameters:

  • dists (torch.Tensor [num_points, num_observations]) – coefficients used in the knn.

  • observations (torch.Tensor [num_observations, dim_observations]) – observation corresponding to each weight dimension.

  • num (int, optional) – number of nearest neighbors. Defaults to 1.

Returns:

KNN

Return type:

torch.Tensor [num_points, dim_observations]

kerch.method.kernel_knn(domain: Tensor, observations: Tensor, num: int = 1, kernel_type: str = 'rbf', **kwargs) → Tensor[source]

For each coefficient, returns the average of the num greatest corresponding kernel values on the domain. The kernel is defined as in kerch.kernel.factory().

Parameters:

  • domain (torch.Tensor [num_observations, dim_domain]) – domain corresponding to each observation.

  • observations (torch.Tensor [num_observations, dim_observations]) – observation corresponding to each domain entry.

  • num (int, optional) – number of nearest neighbors. Defaults to 1.

  • kernel_type (str, optional) – Type of kernel chosen. For the possible choices, please refer to the Factory Type column of the Kernel Module documentation. Defaults to kerch.DEFAULT_KERNEL_TYPE.

  • **kwargs (dict, optional) – Arguments to be passed to the kernel constructor, such as sample or sigma. If an argument is passed that does not exist (e.g. sigma to a linear kernel), it will just be neglected. For the default values, please refer to the default values of the requested kernel.

Returns:

KNN

Return type:

torch.Tensor [num_points, dim_observations]