0.4.1 is released today!
This minor update includes:
CUDA 10.1 support (#950 )
Conda and pip users could install with
conda install -c dglteam dgl-cuda10.1
pip install dgl-cu101
MXNet NN modules support and their examples
- GATConv
- EdgeConv
- SAGEConv
- SGConv
- APPNPConv
- GINConv
- GatedGraphConv
- GMMConv
- ChebConv
- AGNNConv
- NNConv
- DenseGraphConv
- DenseSAGEConv
- DenseChebConv
Miscellaneous
- PyTorch dependency updated to 1.3.0, using BoolTensors in examples (PR #954, @JakeStevens )
- Adapting to API changes in new NetworkX packages (PR #937 @burness , PR #949 @classicsong )
- Bugfix #946