PairwiseDistance¶

class torch.nn.PairwiseDistance(p=2.0, eps=1e-06, keepdim=False)[source]¶

Computes the pairwise distance between input vectors, or between columns of input matrices.

Distances are computed using p-norm, with constant eps added to avoid division by zero if p is negative, i.e.:

\[\mathrm{dist}\left(x, y\right) = \left\Vert x-y + \epsilon e \right\Vert_p, \]

where \(e\) is the vector of ones and the p-norm is given by.

\[\Vert x \Vert _p = \left( \sum_{i=1}^n \vert x_i \vert ^ p \right) ^ {1/p}. \]

Parameters:

p (real, optional) – the norm degree. Can be negative. Default: 2
eps (float, optional) – Small value to avoid division by zero. Default: 1e-6
keepdim (bool, optional) – Determines whether or not to keep the vector dimension. Default: False

Shape:

Input1: \((N, D)\) or \((D)\) where N = batch dimension and D = vector dimension
Input2: \((N, D)\) or \((D)\), same shape as the Input1
Output: \((N)\) or \(()\) based on input dimension. If keepdim is True, then \((N, 1)\) or \((1)\) based on input dimension.

Examples::

>>> pdist = nn.PairwiseDistance(p=2)
>>> input1 = torch.randn(100, 128)
>>> input2 = torch.randn(100, 128)
>>> output = pdist(input1, input2)

Docs