diff --git a/MANIFEST.in b/MANIFEST.in
index bc73983a9c4..ea12b9342b3 100644
--- a/MANIFEST.in
+++ b/MANIFEST.in
@@ -1,2 +1,4 @@
 include python/versioneer.py
-include python/cugraph/_version.py
\ No newline at end of file
+include python/cugraph/_version.py
+include cugraph/experimental/datasets/*.yaml
+include cugraph/experimental/datasets/metadata/*.yaml
\ No newline at end of file
diff --git a/python/cugraph/MANIFEST.in b/python/cugraph/MANIFEST.in
index 51f5db65c13..1f6d9f7a4d0 100644
--- a/python/cugraph/MANIFEST.in
+++ b/python/cugraph/MANIFEST.in
@@ -1,2 +1,4 @@
 include versioneer.py
 include cugraph/_version.py
+include cugraph/experimental/datasets/*.yaml
+include cugraph/experimental/datasets/metadata/*.yaml
\ No newline at end of file
diff --git a/python/cugraph/cugraph/experimental/__init__.py b/python/cugraph/cugraph/experimental/__init__.py
index 9f912346990..964cfe369fe 100644
--- a/python/cugraph/cugraph/experimental/__init__.py
+++ b/python/cugraph/cugraph/experimental/__init__.py
@@ -39,3 +39,5 @@
 find_bicliques = deprecated_warning_wrapper(
     experimental_warning_wrapper(EXPERIMENTAL__find_bicliques)
 )
+
+from cugraph.experimental.datasets.dataset import Dataset
diff --git a/python/cugraph/cugraph/experimental/datasets/__init__.py b/python/cugraph/cugraph/experimental/datasets/__init__.py
new file mode 100644
index 00000000000..fedf9d0dbda
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/__init__.py
@@ -0,0 +1,51 @@
+# Copyright (c) 2022, 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.
+
+
+from cugraph.experimental.datasets.dataset import (
+    Dataset,
+    load_all,
+    set_config,
+    set_download_dir,
+    get_download_dir,
+    default_download_dir
+)
+from cugraph.experimental.datasets import metadata
+from pathlib import Path
+
+
+meta_path = Path(__file__).parent / "metadata"
+
+karate = Dataset(meta_path / "karate.yaml")
+karate_undirected = Dataset(meta_path / "karate_undirected.yaml")
+karate_asymmetric = Dataset(meta_path / "karate_asymmetric.yaml")
+dolphins = Dataset(meta_path / "dolphins.yaml")
+polbooks = Dataset(meta_path / "polbooks.yaml")
+netscience = Dataset(meta_path / "netscience.yaml")
+cyber = Dataset(meta_path / "cyber.yaml")
+small_line = Dataset(meta_path / "small_line.yaml")
+small_tree = Dataset(meta_path / "small_tree.yaml")
+
+
+# LARGE DATASETS
+LARGE_DATASETS = [cyber]
+
+# <10,000 lines
+MEDIUM_DATASETS = [netscience, polbooks]
+
+# <500 lines
+SMALL_DATASETS = [karate, small_line, small_tree, dolphins]
+
+# ALL
+ALL_DATASETS = [karate, dolphins, netscience, polbooks, cyber,
+                small_line, small_tree]
\ No newline at end of file
diff --git a/python/cugraph/cugraph/experimental/datasets/dataset.py b/python/cugraph/cugraph/experimental/datasets/dataset.py
new file mode 100644
index 00000000000..3ae904904f6
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/dataset.py
@@ -0,0 +1,207 @@
+# Copyright (c) 2022, 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 cugraph
+import cudf
+import yaml
+import os
+from pathlib import Path
+
+
+class DefaultDownloadDir:
+    """
+    Maintains the path to the download directory used by Dataset instances.
+    Instances of this class are typically shared by several Dataset instances
+    in order to allow for the download directory to be defined and updated by
+    a single object.
+    """
+    def __init__(self):
+        self._path = Path(os.environ.get("RAPIDS_DATASET_ROOT_DIR",
+                                         Path.home() / ".cugraph/datasets"))
+
+    @property
+    def path(self):
+        """
+        If `path` is not set, set it to the environment variable
+        RAPIDS_DATASET_ROOT_DIR. If the variable is not set, default to the
+        user's home directory.
+        """
+        if self._path is None:
+            self._path = Path(os.environ.get("RAPIDS_DATASET_ROOT_DIR",
+                                             Path.home() /
+                                             ".cugraph/datasets"))
+        return self._path
+
+    @path.setter
+    def path(self, new):
+        self._path = Path(new)
+
+    def clear(self):
+        self._path = None
+
+
+default_download_dir = DefaultDownloadDir()
+
+
+class Dataset:
+    """
+    A Dataset Object, used to easily import edgelist data and cuGraph.Graph
+    instances.
+
+    Parameters
+    ----------
+    meta_data_file_name : yaml file
+        The metadata file for the specific graph dataset, which includes
+        information on the name, type, url link, data loading format, graph
+        properties
+
+    """
+    def __init__(self, meta_data_file_name):
+        with open(meta_data_file_name, 'r') as file:
+            self.metadata = yaml.safe_load(file)
+
+        self._dl_path = default_download_dir
+        self._edgelist = None
+        self._graph = None
+        self._path = None
+
+    def __download_csv(self, url):
+        self._dl_path.path.mkdir(parents=True, exist_ok=True)
+
+        filename = self.metadata['name'] + self.metadata['file_type']
+        if self._dl_path.path.is_dir():
+            df = cudf.read_csv(url)
+            df.to_csv(self._dl_path.path / filename, index=False)
+
+        else:
+            raise RuntimeError(f"The directory {self._dl_path.path.absolute()}"
+                               "does not exist")
+
+    def get_edgelist(self, fetch=False):
+        """
+        Return an Edgelist
+
+        Parameters
+        ----------
+        fetch : Boolean (default=False)
+            Automatically fetch for the dataset from the 'url' location within
+            the YAML file.
+        """
+
+        if self._edgelist is None:
+            full_path = self._dl_path.path / (self.metadata['name'] +
+                                              self.metadata['file_type'])
+
+            if not full_path.is_file():
+                if fetch:
+                    self.__download_csv(self.metadata['url'])
+                else:
+                    raise RuntimeError(f"The datafile {full_path} does not"
+                                       " exist. Try get_edgelist(fetch=True)"
+                                       " to download the datafile")
+
+            self._edgelist = cudf.read_csv(full_path,
+                                           delimiter=self.metadata['delim'],
+                                           names=self.metadata['col_names'],
+                                           dtype=self.metadata['col_types'])
+            self._path = full_path
+
+        return self._edgelist
+
+    def get_graph(self, fetch=False):
+        """
+        Return a Graph object.
+
+        Parameters
+        ----------
+        fetch : Boolean (default=False)
+            Automatically fetch for the dataset from the 'url' location within
+            the YAML file.
+        """
+        if self._edgelist is None:
+            self.get_edgelist(fetch)
+
+        self._graph = cugraph.Graph(directed=self.metadata['is_directed'])
+        self._graph.from_cudf_edgelist(self._edgelist, source='src',
+                                       destination='dst')
+
+        return self._graph
+
+    def get_path(self):
+        """
+        Returns the location of the stored dataset file
+        """
+        if self._path is None:
+            raise RuntimeError("Path to datafile has not been set." +
+                               " Call get_edgelist or get_graph first")
+
+        return self._path.absolute()
+
+
+def load_all(force=False):
+    """
+    Looks in `metadata` directory and fetches all datafiles from the the URLs
+    provided in each YAML file.
+
+    Parameters
+    force : Boolean (default=False)
+        Overwrite any existing copies of datafiles.
+    """
+    default_download_dir.path.mkdir(parents=True, exist_ok=True)
+
+    meta_path = Path(__file__).parent.absolute() / "metadata"
+    for file in meta_path.iterdir():
+        meta = None
+        if file.suffix == '.yaml':
+            with open(meta_path / file, 'r') as metafile:
+                meta = yaml.safe_load(metafile)
+
+            if 'url' in meta:
+                filename = meta['name'] + meta['file_type']
+                save_to = default_download_dir.path / filename
+                if not save_to.is_file() or force:
+                    df = cudf.read_csv(meta['url'])
+                    df.to_csv(save_to, index=False)
+
+
+def set_config(cfgpath):
+    """
+    Read in a custom config file.
+
+    Parameters
+    ----------
+    cfgfile : String
+        Read the custom config file given its path, and override the default
+    """
+    with open(Path(cfgpath), 'r') as file:
+        cfg = yaml.safe_load(file)
+        default_download_dir.path = Path(cfg['download_dir'])
+
+
+def set_download_dir(path):
+    """
+    Set the download directory for fetching datasets
+
+    Parameters
+    ----------
+    path : String
+        Location used to store datafiles
+    """
+    if path is None:
+        default_download_dir.clear()
+    else:
+        default_download_dir.path = path
+
+
+def get_download_dir():
+    return default_download_dir.path.absolute()
diff --git a/python/cugraph/cugraph/experimental/datasets/datasets_config.yaml b/python/cugraph/cugraph/experimental/datasets/datasets_config.yaml
new file mode 100644
index 00000000000..69a79db9cd9
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/datasets_config.yaml
@@ -0,0 +1,5 @@
+---
+fetch: "False"
+force: "False"
+# path where datasets will be downloaded to and stored
+download_dir: "datasets"
diff --git a/python/cugraph/cugraph/experimental/datasets/metadata/cyber.yaml b/python/cugraph/cugraph/experimental/datasets/metadata/cyber.yaml
new file mode 100644
index 00000000000..36858242ec7
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/metadata/cyber.yaml
@@ -0,0 +1,21 @@
+name: cyber
+file_type: .csv
+author: N/A
+url: https://raw.githubusercontent.com/rapidsai/cugraph/branch-22.08/datasets/cyber.csv
+refs: N/A
+col_names:
+  - idx
+  - src
+  - dst
+col_types:
+  - int32
+  - str
+  - str
+delim: ","
+has_loop: true
+is_directed: true
+is_multigraph: false
+is_symmetric: false
+number_of_edges: 54
+number_of_nodes: 314
+number_of_lines: 2546576
diff --git a/python/cugraph/cugraph/experimental/datasets/metadata/dolphins.yaml b/python/cugraph/cugraph/experimental/datasets/metadata/dolphins.yaml
new file mode 100644
index 00000000000..ef07def2b97
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/metadata/dolphins.yaml
@@ -0,0 +1,24 @@
+name: dolphins
+file_type: .csv
+author: D. Lusseau
+url: https://raw.githubusercontent.com/rapidsai/cugraph/branch-22.08/datasets/dolphins.csv
+refs:
+  D. Lusseau, K. Schneider, O. J. Boisseau, P. Haase, E. Slooten, and S. M. Dawson,
+  The bottlenose dolphin community of Doubtful Sound features a large proportion of
+  long-lasting associations, Behavioral Ecology and Sociobiology 54, 396-405 (2003).
+col_names:
+  - src
+  - dst
+  - wgt
+col_types:
+  - int32
+  - int32
+  - float32
+delim: " "
+has_loop: false
+is_directed: true
+is_multigraph: false
+is_symmetric: false
+number_of_edges: 318
+number_of_nodes: 62
+number_of_lines: 318
diff --git a/python/cugraph/cugraph/experimental/datasets/metadata/karate.yaml b/python/cugraph/cugraph/experimental/datasets/metadata/karate.yaml
new file mode 100644
index 00000000000..d86c7b1a241
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/metadata/karate.yaml
@@ -0,0 +1,21 @@
+name: karate-data
+file_type: .csv
+author: Zachary W.
+url: https://raw.githubusercontent.com/rapidsai/cugraph/branch-22.08/datasets/karate-data.csv
+refs:
+  W. W. Zachary, An information flow model for conflict and fission in small groups,
+  Journal of Anthropological Research 33, 452-473 (1977).
+delim: "\t"
+col_names:
+  - src
+  - dst
+col_types:
+  - int32
+  - int32
+has_loop: true
+is_directed: true
+is_multigraph: false
+is_symmetric: true
+number_of_edges: 156
+number_of_nodes: 34
+number_of_lines: 156
diff --git a/python/cugraph/cugraph/experimental/datasets/metadata/karate_asymmetric.yaml b/python/cugraph/cugraph/experimental/datasets/metadata/karate_asymmetric.yaml
new file mode 100644
index 00000000000..b4a81fde29e
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/metadata/karate_asymmetric.yaml
@@ -0,0 +1,23 @@
+name: karate-asymmetric
+file_type: .csv
+author: Zachary W.
+url: https://raw.githubusercontent.com/rapidsai/cugraph/branch-22.08/datasets/karate-asymmetric.csv
+refs:
+  W. W. Zachary, An information flow model for conflict and fission in small groups,
+  Journal of Anthropological Research 33, 452-473 (1977).
+delim: "\t"
+col_names:
+  - src
+  - dst
+  - wgt
+col_types:
+  - int32
+  - int32
+  - float32
+has_loop: true
+is_directed: false
+is_multigraph: false
+is_symmetric: false
+number_of_edges: 78
+number_of_nodes: 34
+number_of_lines: 78
diff --git a/python/cugraph/cugraph/experimental/datasets/metadata/karate_undirected.yaml b/python/cugraph/cugraph/experimental/datasets/metadata/karate_undirected.yaml
new file mode 100644
index 00000000000..061b3361367
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/metadata/karate_undirected.yaml
@@ -0,0 +1,21 @@
+name: karate_undirected
+file_type: .csv
+author: Zachary W.
+url: https://raw.githubusercontent.com/rapidsai/cugraph/branch-22.08/datasets/karate_undirected.csv
+refs:
+  W. W. Zachary, An information flow model for conflict and fission in small groups,
+  Journal of Anthropological Research 33, 452-473 (1977).
+delim: "\t"
+col_names:
+  - src
+  - dst
+col_types:
+  - int32
+  - int32
+has_loop: true
+is_directed: false
+is_multigraph: false
+is_symmetric: true
+number_of_edges: 78
+number_of_nodes: 34
+number_of_lines: 78
diff --git a/python/cugraph/cugraph/experimental/datasets/metadata/netscience.yaml b/python/cugraph/cugraph/experimental/datasets/metadata/netscience.yaml
new file mode 100644
index 00000000000..9c3bd8a6a1d
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/metadata/netscience.yaml
@@ -0,0 +1,21 @@
+name: netscience
+file_type: .csv
+author: Newman, Mark EJ
+url: https://raw.githubusercontent.com/rapidsai/cugraph/branch-22.08/datasets/netscience.csv
+refs: Finding community structure in networks using the eigenvectors of matrices.
+delim: " "
+col_names:
+  - src
+  - dst
+  - wgt
+col_types:
+  - int32
+  - int32
+  - float32
+has_loop: false
+is_directed: true
+is_multigraph: false
+is_symmetric: true
+number_of_edges: 2742
+number_of_nodes: 1461
+number_of_lines: 5484
diff --git a/python/cugraph/cugraph/experimental/datasets/metadata/polbooks.yaml b/python/cugraph/cugraph/experimental/datasets/metadata/polbooks.yaml
new file mode 100644
index 00000000000..75e0e69565d
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/metadata/polbooks.yaml
@@ -0,0 +1,21 @@
+name: polbooks
+file_type: .csv
+author: V. Krebs
+url: https://raw.githubusercontent.com/rapidsai/cugraph/branch-22.08/datasets/polbooks.csv
+refs: null
+delim: " "
+col_names:
+  - src
+  - dst
+  - wgt
+col_types:
+  - int32
+  - int32
+  - float32
+is_directed: true
+has_loop: null
+is_multigraph: null
+is_symmetric: true
+number_of_edges: 882
+number_of_nodes: 105
+number_of_lines: 882
diff --git a/python/cugraph/cugraph/experimental/datasets/metadata/small_line.yaml b/python/cugraph/cugraph/experimental/datasets/metadata/small_line.yaml
new file mode 100644
index 00000000000..9831ff11b30
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/metadata/small_line.yaml
@@ -0,0 +1,21 @@
+name: small_line
+file_type: .csv
+author: null
+url: https://raw.githubusercontent.com/rapidsai/cugraph/branch-22.08/datasets/small_line.csv
+refs: null
+delim: " "
+col_names:
+  - src
+  - dst
+  - wgt
+col_types:
+  - int32
+  - int32
+  - float32
+has_loop: false
+is_directed: false
+is_multigraph: false
+is_symmetric: true
+number_of_edges: 9
+number_of_nodes: 10
+number_of_lines: 8
diff --git a/python/cugraph/cugraph/experimental/datasets/metadata/small_tree.yaml b/python/cugraph/cugraph/experimental/datasets/metadata/small_tree.yaml
new file mode 100644
index 00000000000..942f468c23b
--- /dev/null
+++ b/python/cugraph/cugraph/experimental/datasets/metadata/small_tree.yaml
@@ -0,0 +1,21 @@
+name: small_tree
+file_type: .csv
+author: null
+url: https://raw.githubusercontent.com/rapidsai/cugraph/branch-22.08/datasets/small_tree.csv
+refs: null
+delim: " "
+col_names:
+  - src
+  - dst
+  - wgt
+col_types:
+  - int32
+  - int32
+  - float32
+has_loop: false
+is_directed: true
+is_multigraph: false
+is_symmetric: true
+number_of_edges: 11
+number_of_nodes: 9
+number_of_lines: 11
diff --git a/python/cugraph/cugraph/tests/test_dataset.py b/python/cugraph/cugraph/tests/test_dataset.py
new file mode 100644
index 00000000000..093f1382bcb
--- /dev/null
+++ b/python/cugraph/cugraph/tests/test_dataset.py
@@ -0,0 +1,169 @@
+# Copyright (c) 2022, 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 pytest
+import yaml
+import os
+from pathlib import Path
+from tempfile import NamedTemporaryFile, TemporaryDirectory
+from cugraph.experimental.datasets import (ALL_DATASETS)
+
+
+# =============================================================================
+# Pytest Setup / Teardown - called for each test function
+# =============================================================================
+
+# Use this to simulate a fresh API import
+@pytest.fixture
+def datasets():
+    from cugraph.experimental import datasets
+    yield datasets
+    del datasets
+    clear_locals()
+
+
+def clear_locals():
+    for dataset in ALL_DATASETS:
+        dataset._edgelist = None
+        dataset._graph = None
+        dataset._path = None
+
+
+# We use this to create tempfiles that act as config files when we call
+# set_config(). Arguments passed will act as custom download directories
+def create_config(custom_path="custom_storage_location"):
+    config_yaml = """
+                    fetch: False
+                    force: False
+                    download_dir: None
+                    """
+    c = yaml.safe_load(config_yaml)
+    c['download_dir'] = custom_path
+
+    outfile = NamedTemporaryFile()
+    with open(outfile.name, 'w') as f:
+        yaml.dump(c, f, sort_keys=False)
+
+    return outfile
+
+
+# setting download_dir to None effectively re-initialized the default
+def test_env_var(datasets):
+    os.environ['RAPIDS_DATASET_ROOT_DIR'] = 'custom_storage_location'
+    datasets.set_download_dir(None)
+
+    expected_path = Path("custom_storage_location").absolute()
+    assert datasets.get_download_dir() == expected_path
+
+    del os.environ['RAPIDS_DATASET_ROOT_DIR']
+
+
+def test_home_dir(datasets):
+    datasets.set_download_dir(None)
+    expected_path = Path.home() / ".cugraph/datasets"
+
+    assert datasets.get_download_dir() == expected_path
+
+
+def test_set_config(datasets):
+    cfg = create_config()
+    datasets.set_config(cfg.name)
+
+    assert datasets.get_download_dir() == \
+           Path("custom_storage_location").absolute()
+
+    cfg.close()
+
+
+def test_set_download_dir(datasets):
+    tmpd = TemporaryDirectory()
+    datasets.set_download_dir(tmpd.name)
+
+    assert datasets.get_download_dir() == Path(tmpd.name).absolute()
+
+    tmpd.cleanup()
+
+
+def test_load_all(datasets):
+    tmpd = TemporaryDirectory()
+    cfg = create_config(custom_path=tmpd.name)
+    datasets.set_config(cfg.name)
+    datasets.load_all()
+
+    for data in datasets.ALL_DATASETS:
+        file_path = Path(tmpd.name) / (data.metadata['name'] +
+                                       data.metadata['file_type'])
+        assert file_path.is_file()
+
+    tmpd.cleanup()
+
+
+@pytest.mark.parametrize("dataset", ALL_DATASETS)
+def test_fetch(dataset, datasets):
+    tmpd = TemporaryDirectory()
+    cfg = create_config(custom_path=tmpd.name)
+    datasets.set_config(cfg.name)
+
+    E = dataset.get_edgelist(fetch=True)
+
+    assert E is not None
+    assert dataset.get_path().is_file()
+
+    tmpd.cleanup()
+
+
+@pytest.mark.parametrize("dataset", ALL_DATASETS)
+def test_get_edgelist(dataset, datasets):
+    tmpd = TemporaryDirectory()
+    datasets.set_download_dir(tmpd.name)
+    E = dataset.get_edgelist(fetch=True)
+
+    assert E is not None
+
+    tmpd.cleanup()
+
+
+@pytest.mark.parametrize("dataset", ALL_DATASETS)
+def test_get_graph(dataset, datasets):
+    tmpd = TemporaryDirectory()
+    datasets.set_download_dir(tmpd.name)
+    G = dataset.get_graph(fetch=True)
+
+    assert G is not None
+
+    tmpd.cleanup()
+
+
+@pytest.mark.parametrize("dataset", ALL_DATASETS)
+def test_metadata(dataset):
+    M = dataset.metadata
+
+    assert M is not None
+
+
+@pytest.mark.parametrize("dataset", ALL_DATASETS)
+def test_get_path(dataset, datasets):
+    tmpd = TemporaryDirectory()
+    datasets.set_download_dir(tmpd.name)
+    dataset.get_edgelist(fetch=True)
+
+    assert dataset.get_path().is_file()
+    tmpd.cleanup()
+
+
+# Path is None until a dataset initializes its edgelist
+@pytest.mark.parametrize("dataset", ALL_DATASETS)
+def test_get_path_raises(dataset):
+    with pytest.raises(RuntimeError):
+        dataset.get_path()
diff --git a/python/cugraph/setup.py b/python/cugraph/setup.py
index 91bdea4a2bc..c2d994969c0 100644
--- a/python/cugraph/setup.py
+++ b/python/cugraph/setup.py
@@ -191,6 +191,11 @@ def finalize_options(self):
       setup_requires=['Cython>=0.29,<0.30'],
       ext_modules=extensions,
       packages=find_packages(include=['cugraph', 'cugraph.*']),
+      include_package_data=True,
+      package_data={
+          '': ['python/cugraph/cugraph/experimental/datasets/metadata/*.yaml',
+               'python/cugraph/cugraph/experimental/datasets/*.yaml'],
+      },
       install_requires=INSTALL_REQUIRES,
       license="Apache",
       cmdclass=cmdclass,