Google I/O is a wrap! Catch up on TensorFlow sessions View sessions


Build the Laplacian matrix of a nearest neighbor graph.

Given input embedding vectors, this utility returns the Laplacian matrix of the induced k-nearest-neighbor graph.

input_vecs a Tensor. Input embedding vectors (one per row). Shaped [num_vectors, ...].
k an integer. Number of nearest neighbors to use.

The graph Laplacian matrix. A dense float Tensor of shape [num_vectors, num_vectors], where num_vectors is the number of input embedding vectors (Tensor).