spektral

spektral copied to clipboard

Loading a saved model has different structure. Error to use the loaded model.

3

Hello, I have a model like this: ``` l2_reg = 5e-4 # Regularization rate for l2 class MyFirstGNN(Model): def __init__(self, n_labels): super().__init__() self.conv1 = GCNConv(32, activation="elu", kernel_regularizer=l2(l2_reg)) self.conv2 = GCNConv(32,...

JaeLee18

GATConv compatibility with Disjoint mode?

9

Dear Experts: I was trying to use GATConv in disjoint mode with disjoint data loader. But when I run model.fit, I got the following error that I couldn't figure out...

edshui

Training a network with heterogeneous data

4

Hi! I'm trying to build a regression model which can predict a parameter starting from a dataset in which the X is represented by a couple (Graph, integer parameter) and...

AndywinXp

How import the egde features tensor into GCNConv layer?

1

the GCNConv layer gets two inputs that node with features and weighted adjacency matrix, but how we import the egde features?

Lance729

Anaconda Support

3

This is more of a request than a functional issue. It would be extremely nice to be able to install and maintain spektral from conda. pip install into a conda...

Omnipius

[Feature request] More layers that support directed and weighted graphs

1

I took a look at the current layers, and the ones that support directed graphs only accept binary adjacency matrix as input (DiffusionConv is an exception but it's not performing...

claudiocapanema

[Feature Request] Could you please add the implementation of 3D-GCN? 

1

'Convolution in the Cloud: Learning Deformable Kernels in 3D Graph Convolution Networks for Point Cloud Analysis' is a good paper with 3D-GCN, whose funciton contain graph classification and node calssification....

WonderSimiliar

[Feature Request] GraphSAGE model support for edge features

3

Hi, As I understand, the current GraphSAGE model takes into account the node level features for model training. Is there any example / support for GraphSAGE model which also supports...

hmpatel

Graculus pooling and how to pool the easiest way?

1

Hi @danielegrattarola , first of all great work. Do you have graculus pooling implemented (the one from Defferard's original ChebNet paper)? I could not find it, but it might be...

Matimath

Implementation of Eigen and laplacian Pooling

1

Hi Daniele, Fantastic work! Can you add those two pooling methods please ? https://arxiv.org/pdf/1905.11577.pdf & https://arxiv.org/pdf/1904.13107.pdf (I have a source code but not very functional) thanks in advance, Guillaume

thegodone