Skip to content

Commit

Permalink
fix channel, use new location for dgl wheels
Browse files Browse the repository at this point in the history
  • Loading branch information
jameslamb committed Oct 4, 2024
1 parent f880686 commit 317b86b
Show file tree
Hide file tree
Showing 3 changed files with 5 additions and 5 deletions.
3 changes: 1 addition & 2 deletions ci/test_python.sh
Original file line number Diff line number Diff line change
Expand Up @@ -153,14 +153,13 @@ if [[ "${RAPIDS_CUDA_VERSION}" == "11.8.0" ]]; then
--channel "${CPP_CHANNEL}" \
--channel "${PYTHON_CHANNEL}" \
--channel conda-forge \
--channel dglteam/label/cu118 \
--channel dglteam/label/th23_cu118 \
--channel nvidia \
"libcugraph=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
"pylibcugraph=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
"pylibcugraphops=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
"cugraph=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
"cugraph-dgl=${RAPIDS_VERSION_MAJOR_MINOR}.*" \
'dgl>=1.1.0.cu*,<=2.0.0.cu*' \
'pytorch>=2.3,<2.4' \
'cuda-version=11.8'

Expand Down
4 changes: 2 additions & 2 deletions ci/test_wheel_cugraph-dgl.sh
Original file line number Diff line number Diff line change
Expand Up @@ -30,10 +30,10 @@ else
PYTORCH_CUDA_VER=$PKG_CUDA_VER
fi
PYTORCH_URL="https://download.pytorch.org/whl/cu${PYTORCH_CUDA_VER}"
DGL_URL="https://data.dgl.ai/wheels/cu${PYTORCH_CUDA_VER}/repo.html"
DGL_URL="https://data.dgl.ai/wheels/torch-2.3/cu${PYTORCH_CUDA_VER}/repo.html"

rapids-logger "Installing PyTorch and DGL"
rapids-retry python -m pip install torch==2.3.0 --index-url ${PYTORCH_URL}
rapids-retry python -m pip install dgl==2.2.1 --find-links ${DGL_URL}
rapids-retry python -m pip install dgl==2.4.0 --find-links ${DGL_URL}

python -m pytest python/cugraph-dgl/tests
3 changes: 2 additions & 1 deletion docs/cugraph/source/wholegraph/installation/container.md
Original file line number Diff line number Diff line change
Expand Up @@ -24,6 +24,7 @@ RUN pip3 install Cython setuputils3 scikit-build nanobind pytest-forked pytest
To run GNN applications, you may also need cuGraphOps, DGL and/or PyG libraries to run the GNN layers.
You may refer to [DGL](https://www.dgl.ai/pages/start.html) or [PyG](https://pytorch-geometric.readthedocs.io/en/latest/notes/installation.html)
For example, to install DGL, you may need to add:

```dockerfile
RUN pip3 install dgl -f https://data.dgl.ai/wheels/cu118/repo.html
RUN pip3 install dgl -f https://data.dgl.ai/wheels/torch-2.3/cu118/repo.html
```

0 comments on commit 317b86b

Please sign in to comment.