GNN are a special class of neural networks that are capable of working with data that is represented in graph form. These networks are heavily motivated by Convolutional Neural Networks and graph embedding. CNN’s are not able to handle graph data because the nodes in the graphs aren’t represented in any order and the fact that dependency information between two nodes is represented by edges.
Questions tagged [gnn]
56 questions
0
votes
0 answers
How to mask nodes in PyTorch Geometric for a particular layer?
I am working on a node classification problem, and I have a graph where some nodes do not have features, and each of these nodes are connected to several nodes with features.
I want to aggregate the features from the nodes with features to the nodes…
Ralff
- 401
- 2
- 5
- 15
0
votes
1 answer
Using a target size (torch.Size([80670, 1])) that is different to the input size (torch.Size([80670]))
I want to implement GCN with my own dataset (from Pytorch geometric tutorial). I copied my notebook below.
It is working when I use the original implementation, However, if I want to use custom data then it gives the below error:
My…
wasif khan
- 27
- 6
0
votes
0 answers
Who can help me to understand the "torch.einsum('ijh,jhf->ihf', a, g)"?
When I followed the tutorial of an implementation about GAT(Graph Attention Network) here. At the line 191: attn_res = torch.einsum('ijh,jhf->ihf', a, g) , I don't konw how the rule ijh,jhf->ihf works. Then I went to document of Pytorch, I noticed a…
guoqing qin
- 9
- 1
0
votes
0 answers
What does the scatter_max() operation do in the softmax() function of PyG?
So the main logic inside pyg.utils.softmax() is as following:
N = maybe_num_nodes(index, num_nodes)
src_max = scatter(src, index, dim, dim_size=N, reduce='max')
src_max = src_max.index_select(dim, index)
out = (src - src_max).exp()
out_sum =…
ht.lee
- 23
- 5
0
votes
1 answer
Repetitive word predictions in RNN
Hello dear community,
I am training a Seq2Seq model to generate a question based on a graph. Both train and val loss are converging, but the generated questions (on either train or test set) are nonsense and contain mostly repetition of tokens. I…
DustyAvocado
- 43
- 4
0
votes
0 answers
How to import the OGB dataset?
I've installed the ogb packages, but it doesn't work. when i run the code, the process will show nothing, only waiting........, then i mark the ogb with 2 'print' , then it turns out like below
the mark...
the code would stay there forever
pic
Eric
- 1
- 1
0
votes
1 answer
Colab Session Crashes on importing torch_geometric.data
My colab session always crashes while trying to import torch_geometric.data module. For reference, the code I am writing is as…
unstableEquilibrium
- 157
- 1
- 7
0
votes
1 answer
cannot find pysat=0.1.3 dependency
I am getting error as
from pysat.solvers import Glucose3
ModuleNotFoundError: No module named 'pysat.solvers'*
when I am trying newer version of pysat.
I cannot find the required older version that is pysat=0.1.3.
-1
votes
0 answers
Single feature explanation in GAT
I have trained a GNN using GATv2 layers in PyTorch and am looking to explain my model. I have used AttentionExplainer to obtain edge weights of the trained model but I am also looking to get the contributions of each of the features in my node…
kristina kordova
- 19
- 1
-3
votes
0 answers
Pre-trained Link prediction on ConceptNet
I'm looking for a pre-trained link predictor which trained on ConceptNet, do you have any suggestion? Also if you know any other link predictor it's work for me!
Arshia Hemmat
- 15
- 2