Simple Linear layer with dropout.

Simple GCN layer.

Hyperbolic activation layer.

Hyperbolic linear layer.

Euclidean Manifold class.

Fermi Dirac to compute edge probabilities based on distances.

Hyperbolic aggregation layer.

Sparse version GAT layer, similar to https://arxiv.org/abs/1710.10903

Base model for graph embedding tasks.

Decoder abstract class for node classification tasks.

Encoder abstract class.

Graph Attention Networks.

Graph Attention Decoder.

Graph Convolution Networks.

Graph Convolution Decoder.

Hyperbolic-GCN.

Hyperbolic Neural Networks.

Hyperbolic neural networks layer.

Hyperbolic graph convolution layer.

Hyperboloid manifold class. We use the following convention: -x0^2 + x1^2 + ... + xd^2 = -K c = 1 / K is the hyperbolic curvature.

Base model for link prediction task.

MLP Decoder for Hyperbolic/Euclidean node classification models.

Multi-layer perceptron.

Abstract class to define operations on a manifold.

Subclass of torch.nn.Parameter for Riemannian optimization.

Base model for node classification task.

PoicareBall Manifold class. We use the following convention: x0^2 + x1^2 + ... + xd^2 < 1 / c Note that 1/sqrt(c) is the Poincare ball

r"""Riemannian Adam with the same API as :class:`torch.optim.Adam` Parameters ---------- params : iterable iterable of parameters

Shallow Embedding method. Learns embeddings or loads pretrained embeddings and uses an MLP for classification.

Special function for only sparse region backpropataion layer.