Cluster the dataset using affinity propagation.

Options

• :iterations (pos_integer/0) - Number of iterations of the algorithm. The default value is 300.

• :damping_factor (float/0) - Damping factor in the range [0.5, 1.0) is the extent to which the current value is maintained relative to incoming values (weighted 1 - damping). The default value is 0.5.

• :self_preference - Self preference.

• :key - Determines random number generation for centroid initialization. If the key is not provided, it is set to Nx.Random.key(System.system_time()).

• :learning_loop_unroll (boolean/0) - If true, the learning loop is unrolled. The default value is false.

Return Values

The function returns a struct with the following parameters:

• :affinity_matrix - Affinity matrix. It is a negated squared euclidean distance of each pair of points.

• :clusters_centers - Cluster centers from the initial data.

• :cluster_centers_indices - Indices of cluster centers.

• :num_clusters - Number of clusters.

Examples

iex> key = Nx.Random.key(42)
iex> x = Nx.tensor([[12,5,78,2], [1,-5,7,32], [-1,3,6,1], [1,-2,5,2]])
iex> Scholar.Cluster.AffinityPropagation.fit(x, key: key)
%Scholar.Cluster.AffinityPropagation{
  labels: Nx.tensor([0, 3, 3, 3]),
  cluster_centers_indices: Nx.tensor([0, -1, -1, 3]),
  affinity_matrix: Nx.tensor(
    [
      [-0.0, -6162.0, -5358.0, -5499.0],
      [-6162.0, -0.0, -1030.0, -913.0],
      [-5358.0, -1030.0, -0.0, -31.0],
      [-5499.0, -913.0, -31.0, -0.0]
    ]),
  cluster_centers: Nx.tensor(
    [
      [12.0, 5.0, 78.0, 2.0],
      [:infinity, :infinity, :infinity, :infinity],
      [:infinity, :infinity, :infinity, :infinity],
      [1.0, -2.0, 5.0, 2.0]
    ]
  ),
  num_clusters: Nx.tensor(2, type: :u64)
}

Predict the closest cluster each sample in x belongs to.

Examples

iex> key = Nx.Random.key(42)
iex> x = Nx.tensor([[12,5,78,2], [1,5,7,32], [1,3,6,1], [1,2,5,2]])
iex> model = Scholar.Cluster.AffinityPropagation.fit(x, key: key)
iex> model = Scholar.Cluster.AffinityPropagation.prune(model)
iex> Scholar.Cluster.AffinityPropagation.predict(model, Nx.tensor([[1,6,2,6], [8,3,8,2]]))
#Nx.Tensor<
  s64[2]
  [1, 1]
>

Optionally prune clusters, indices, and labels to only valid entries.

It returns an updated and pruned model.

Examples

iex> key = Nx.Random.key(42)
iex> x = Nx.tensor([[12,5,78,2], [1,-5,7,32], [-1,3,6,1], [1,-2,5,2]])
iex> model = Scholar.Cluster.AffinityPropagation.fit(x, key: key)
iex> Scholar.Cluster.AffinityPropagation.prune(model)
%Scholar.Cluster.AffinityPropagation{
  labels: Nx.tensor([0, 1, 1, 1]),
  cluster_centers_indices: Nx.tensor([0, 3]),
  affinity_matrix: Nx.tensor(
    [
      [-0.0, -6162.0, -5358.0, -5499.0],
      [-6162.0, -0.0, -1030.0, -913.0],
      [-5358.0, -1030.0, -0.0, -31.0],
      [-5499.0, -913.0, -31.0, -0.0]
    ]),
  cluster_centers: Nx.tensor(
    [
      [12.0, 5.0, 78.0, 2.0],
      [1.0, -2.0, 5.0, 2.0]
    ]
  ),
  num_clusters: Nx.tensor(2, type: :u64)
}