layer_diffuse_features: DiffuseFeatures

In rdinnager/rspektral: What the Package Does (One Line, Title Case)

Description

\loadmathjax

Utility layer calculating a single channel of the diffusional convolution.

Procedure is based on https://arxiv.org/abs/1707.01926

Input

• Node features of shape ([batch], N, F);

• Normalized adjacency or attention coef. matrix \mjeqn\hat A of shape ([batch], N, N); Use DiffusionConvolution.preprocess to normalize.

Output

• Node features with the same shape as the input, but with the last dimension changed to \mjeqn1.

Usage

layer_diffuse_features(
  object,
  num_diffusion_steps,
  kernel_initializer,
  kernel_regularizer,
  kernel_constraint,
  ...
)

Arguments

num_diffusion_steps	How many diffusion steps to consider. \(K\) in paper.
kernel_initializer	initializer for the weights
kernel_regularizer	regularization applied to the kernel vectors
kernel_constraint	constraint applied to the kernel vectors

rdinnager/rspektral documentation built on June 12, 2021, 1:26 a.m.