About this deal
The graph convolutional operator with initial residual connections and identity mapping (GCNII) from the "Simple and Deep Graph Convolutional Networks" paper. Creates a criterion that optimizes a two-class classification logistic loss between input tensor x x x and target tensor y y y (containing 1 or -1). The dynamic edge convolutional operator from the "Dynamic Graph CNN for Learning on Point Clouds" paper (see torch_geometric.
Applies batch normalization over a batch of heterogeneous features as described in the "Batch Normalization: Accelerating Deep Network Training by Reducing Internal Covariate Shift" paper. The label propagation operator, firstly introduced in the "Learning from Labeled and Unlabeled Data with Label Propagation" paper. The (translation-invariant) feature-steered convolutional operator from the "FeaStNet: Feature-Steered Graph Convolutions for 3D Shape Analysis" paper. The MinCut pooling operator from the "Spectral Clustering in Graph Neural Networks for Graph Pooling" paper. The LightGCN model from the "LightGCN: Simplifying and Powering Graph Convolution Network for Recommendation" paper.
The LINKX model from the "Large Scale Learning on Non-Homophilous Graphs: New Benchmarks and Strong Simple Methods" paper. Applies Instance Normalization over a 2D (unbatched) or 3D (batched) input as described in the paper Instance Normalization: The Missing Ingredient for Fast Stylization.
Applies the gated linear unit function G L U ( a , b ) = a ⊗ σ ( b ) {GLU}(a, b)= a \otimes \sigma(b) G LU ( a , b ) = a ⊗ σ ( b ) where a a a is the first half of the input matrices and b b b is the second half.The graph convolutional operator from the "Semi-supervised Classification with Graph Convolutional Networks" paper. Convert the output of Captum attribution methods which is a tuple of attributions to two dictionaries with node and edge attribution tensors. Creates a criterion that optimizes a multi-class multi-classification hinge loss (margin-based loss) between input x x x (a 2D mini-batch Tensor) and output y y y (which is a 2D Tensor of target class indices). The pathfinder discovery network convolutional operator from the "Pathfinder Discovery Networks for Neural Message Passing" paper.
InstanceNorm1d module with lazy initialization of the num_features argument of the InstanceNorm1d that is inferred from the input. Aggregation functions play an important role in the message passing framework and the readout functions of Graph Neural Networks. The differentiable group normalization layer from the "Towards Deeper Graph Neural Networks with Differentiable Group Normalization" paper, which normalizes node features group-wise via a learnable soft cluster assignment. The fused graph attention operator from the "Understanding GNN Computational Graph: A Coordinated Computation, IO, and Memory Perspective" paper.
Performs LSTM-style aggregation in which the elements to aggregate are interpreted as a sequence, as described in the "Inductive Representation Learning on Large Graphs" paper. The path integral based convolutional operator from the "Path Integral Based Convolution and Pooling for Graph Neural Networks" paper.