docarray.math.distance package

Submodules

• docarray.math.distance.numpy module
  • cosine()
  • sqeuclidean()
  • sparse_cosine()
  • sparse_sqeuclidean()
  • sparse_euclidean()
  • euclidean()
• docarray.math.distance.paddle module
  • cosine()
  • sqeuclidean()
  • euclidean()
• docarray.math.distance.tensorflow module
  • cosine()
  • sqeuclidean()
  • euclidean()
• docarray.math.distance.torch module
  • cosine()
  • euclidean()
  • sqeuclidean()

Module contents

docarray.math.distance.pdist(x_mat, metric)

Computes Pairwise distances between observations in n-dimensional space.

Parameters:

• x_mat (ArrayType) – Union[‘np.ndarray’,’scipy.sparse.csr_matrix’, ‘scipy.sparse.coo_matrix’] of ndim 2

• metric (str) – string describing the metric type

Return type:

np.ndarray

Returns:

np.ndarray of ndim 2

docarray.math.distance.cdist(x_mat, y_mat, metric, device='cpu')

Computes the pairwise distance between each row of X and each row on Y according to metric. - Let n_x = x_mat.shape[0] - Let n_y = y_mat.shape[0] - Returns a matrix dist of shape (n_x, n_y) with dist[i,j] = metric(x_mat[i], y_mat[j]). :type x_mat: ArrayType :param x_mat: numpy or scipy array of ndim 2 :type y_mat: ArrayType :param y_mat: numpy or scipy array of ndim 2 :type metric: str :param metric: string describing the metric type :type device: str :param device: the computational device, can be either cpu or cuda. :rtype: np.ndarray :return: np.ndarray of ndim 2