Add CI to cugraph_dgl (#3312)

This PR replaces #3293 .

Authors:
  - Vibhu Jawa (https://github.com/VibhuJawa)

Approvers:
  - Alex Barghi (https://github.com/alexbarghi-nv)
  - AJ Schmidt (https://github.com/ajschmidt8)
  - Rick Ratzel (https://github.com/rlratzel)

URL: #3312

ci/test_python.sh

@@ -91,6 +91,60 @@ pytest \
   cugraph/pytest-based/bench_algos.py
 popd
 
+if [[ "${RAPIDS_CUDA_VERSION}" == "11.8.0" ]]; then
+  if [[ "${RUNNER_ARCH}" != "ARM64" ]]; then
+    # we are only testing in a single cuda version
+    # because of pytorch and rapids compatibilty problems
+    rapids-mamba-retry env create --force -f env.yaml -n test_cugraph_dgl
+
+    # activate test_cugraph_dgl environment for dgl
+    set +u
+    conda activate test_cugraph_dgl
+    set -u
+    rapids-mamba-retry install \
+      --channel "${CPP_CHANNEL}" \
+      --channel "${PYTHON_CHANNEL}" \
+      --channel pytorch \
+      --channel pytorch-nightly \
+      --channel dglteam/label/cu117 \
+      --channel nvidia \
+      libcugraph \
+      pylibcugraph \
+      cugraph \
+      cugraph-dgl \
+      'dgl>=1.0' \
+      'pytorch>=2.0' \
+      'pytorch-cuda>=11.8'
+
+    rapids-print-env
+
+    rapids-logger "pytest cugraph_dgl (single GPU)"
+    pushd python/cugraph-dgl/tests
+    pytest \
+      --cache-clear \
+      --ignore=mg \
+      --ignore=nn \
+      --junitxml="${RAPIDS_TESTS_DIR}/junit-cugraph-dgl.xml" \
+      --cov-config=../../.coveragerc \
+      --cov=cugraph_dgl \
+      --cov-report=xml:"${RAPIDS_COVERAGE_DIR}/cugraph-dgl-coverage.xml" \
+      --cov-report=term \
+      .
+    popd
+
+    # Reactivate the test environment back
+    set +u
+    conda deactivate
+    conda activate test
+    set -u
+  else
+    rapids-logger "skipping cugraph_dgl pytest on ARM64"
+  fi
+else
+  rapids-logger "skipping cugraph_dgl pytest on CUDA!=11.8"
+fi
+
+
 rapids-logger "pytest cugraph_pyg (single GPU)"
 pushd python/cugraph-pyg/cugraph_pyg
 # rmat is not tested because of multi-GPU testing

conda/environments/all_cuda-118_arch-x86_64.yaml

@@ -37,7 +37,6 @@ dependencies:
 - notebook>=0.5.0
 - numpydoc
 - nvcc_linux-64=11.8
-- ogb
 - openmpi
 - pip
 - pre-commit

dependencies.yaml

@@ -203,7 +203,6 @@ dependencies:
     common:
       - output_types: [conda, requirements]
         packages:
-          - ogb
           - py
           - pytest
           - pytest-cov

python/cugraph-dgl/cugraph_dgl/dataloading/utils/sampling_helpers.py

@@ -21,6 +21,11 @@
 
 
 def cast_to_tensor(ser: cudf.Series):
+    if len(ser) == 0:
+        # Empty series can not be converted to pytorch cuda tensor
+        t = torch.from_numpy(ser.values.get())
+        return t.to("cuda")
+
     return torch.as_tensor(ser.values, device="cuda")
 
 

python/cugraph-dgl/tests/test_utils.py

@@ -0,0 +1,25 @@
+# Copyright (c) 2022-2023, NVIDIA CORPORATION.
+# Licensed under the Apache License, Version 2.0 (the "License");
+# you may not use this file except in compliance with the License.
+# You may obtain a copy of the License at
+#
+#     http://www.apache.org/licenses/LICENSE-2.0
+#
+# Unless required by applicable law or agreed to in writing, software
+# distributed under the License is distributed on an "AS IS" BASIS,
+# WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
+# See the License for the specific language governing permissions and
+# limitations under the License.
+
+import cudf
+import cupy as cp
+import torch
+
+from cugraph_dgl.dataloading.utils.sampling_helpers import cast_to_tensor
+
+
+def test_casting_empty_array():
+    ar = cp.zeros(shape=0, dtype=cp.int32)
+    ser = cudf.Series(ar)
+    output_tensor = cast_to_tensor(ser)
+    assert output_tensor.dtype == torch.int32