| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
|
[Release][Blog] Accelerating Partitioning of Billion-scale Graphs with DGL v0.9.1
Graphs is ubiquitous to represent relational data, and many real-world applications such as recommendation and fraud detection involve learning from massive graphs. As such, GNNs has emerged as a powerful family of model… |
|
2 | 20 | September 20, 2022 |
|
[Release] v0.9 Release Highlights
Six years after the first Graph Convolutional Networks paper, researchers are actively investigating more advanced GNN architecture or training methodology. As the developer team of DGL, we closely watch those new resear… |
|
2 | 131 | August 3, 2022 |
|
Frequently Asked Questions (FAQ)
Are DGLGraphs directed or not? How to represent an undirected graph? All DGLGraphs are directed. To represent an undirected graph, you need to create edges for both directions. dgl.to_bidirected can be helpful, which… |
|
2 | 5569 | January 13, 2021 |
| DGL load_graphs being terribly slow compared to PYG, why? |
|
0 | 3 | September 24, 2022 |
| NeighborSampler depth |
|
0 | 13 | September 21, 2022 |
| [SOLVED] Dataloader returns only some edge weights |
|
2 | 30 | September 19, 2022 |
| Captum.ai Explainability for Heterogeneous Graphs |
|
10 | 106 | September 19, 2022 |
| Why need to convert input graph to homogenous graph in DistDGL |
|
1 | 25 | September 19, 2022 |
| Graphsage C++ implementation |
|
1 | 23 | September 19, 2022 |
| A potential bug in DGL examples |
|
1 | 31 | September 18, 2022 |
| Training Large Graph for edge regression |
|
4 | 36 | September 17, 2022 |
| CUDA kernel launch error: no kernel image is available for execution on the device |
|
3 | 22 | September 17, 2022 |
| Question about the iteration over DGL Dataloader |
|
1 | 27 | September 17, 2022 |
| Edge-level representation learning |
|
2 | 24 | September 17, 2022 |
| Heterogeneous to bipartite graph dgl |
|
2 | 32 | September 13, 2022 |
| Mask edges for RGCN |
|
3 | 52 | September 13, 2022 |
| MovieLens - Hetrogenous Graph | GraphSAGE | Link Prediction |
|
1 | 78 | September 13, 2022 |
| Identifying consecutive edges |
|
1 | 35 | September 10, 2022 |
| Ignore edge type when sampling neighbours |
|
4 | 48 | September 8, 2022 |
| Adjacency matrix of a block is not symmetric |
|
1 | 44 | September 8, 2022 |
| Negative edges labels for link prediction |
|
2 | 57 | September 7, 2022 |
| Dataloading efficiency bottleneck at to_block |
|
1 | 35 | September 7, 2022 |
| DGNConv is not working without eig_vec |
|
1 | 52 | September 6, 2022 |
| V2: Implementing LADIES sampler |
|
2 | 50 | September 6, 2022 |
| Convolution on 'edge' in case of heterogeneous graph |
|
6 | 51 | September 6, 2022 |
| Difference between send(), recv() and update_all()? |
|
2 | 31 | September 6, 2022 |
| RuntimeError: the derivative for 'unique_dim' is not implemented |
|
3 | 43 | September 5, 2022 |
| Difficulties with layer based sampling implementation (LADIES sampler) |
|
8 | 71 | September 5, 2022 |
| How to measure the CPU neighborhood sampling time? |
|
1 | 41 | September 5, 2022 |
| Graph-level data? |
|
2 | 53 | September 5, 2022 |
