Pairwise distance — nn_pairwise_distance • torch

Computes the batchwise pairwise distance between vectors $v_{1}$ , $v_{2}$ using the p-norm:

Usage

nn_pairwise_distance(p = 2, eps = 1e-06, keepdim = FALSE)

Arguments

p: (real): the norm degree. 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

Details

$‖ x ‖_{p} = {(\sum_{i = 1}^{n} | x_{i} |^{p})}^{1 / p} .$

Shape

Input1: $(N, D)$ where D = vector dimension
Input2: $(N, D)$ , same shape as the Input1
Output: $(N)$ . If keepdim is TRUE, then $(N, 1)$ .

Examples

if (torch_is_installed()) {
pdist <- nn_pairwise_distance(p = 2)
input1 <- torch_randn(100, 128)
input2 <- torch_randn(100, 128)
output <- pdist(input1, input2)
}