Skip to content

Add sampled input to DistNeighborSampler._sample_from_nodes #160

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Open
wants to merge 2 commits into
base: main
Choose a base branch
from
Open

Add sampled input to DistNeighborSampler._sample_from_nodes #160

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
ae858d4
Merge pull request #1 from alibaba/main
kmontemayor2-sc Apr 21, 2025
3850871
Add input seeds to sampling output for node sampler
kmontemayor2-sc May 5, 2025
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions graphlearn_torch/python/distributed/dist_neighbor_sampler.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -358,7 +358,7 @@ async def _sample_from_nodes(
num_sampled_nodes=num_sampled_nodes,
num_sampled_edges=num_sampled_edges,
input_type=input_type,
metadata={}
metadata={'input_seeds': input_seeds},
)

else:
Expand Down Expand Up @@ -389,7 +389,7 @@ async def _sample_from_nodes(
batch=batch,
num_sampled_nodes=num_sampled_nodes,
num_sampled_edges=num_sampled_edges,
metadata={}
metadata={'input_seeds': input_seeds},
)
# Reclaim inducer into pool.
self.inducer_pool.put(inducer)
Expand Down
29 changes: 16 additions & 13 deletions graphlearn_torch/python/loader/transform.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -113,21 +113,24 @@ def to_hetero_data(
# update meta data
input_type = hetero_sampler_out.input_type
if isinstance(hetero_sampler_out.metadata, dict):
# if edge_dir == 'out', we need to reverse the edge type
res_edge_type = reverse_edge_type(input_type) if edge_dir == 'out' else input_type
for k, v in hetero_sampler_out.metadata.items():
if k == 'edge_label_index':
if edge_dir == 'out':
data[res_edge_type]['edge_label_index'] = \
torch.stack((v[1], v[0]), dim=0)
if isinstance(input_type, tuple):
Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Making this change as input_type can also be a NodeType or None, both of which will throw if passed into reverse_edge_type.

# if edge_dir == 'out', we need to reverse the edge type
res_edge_type = reverse_edge_type(input_type) if edge_dir == 'out' else input_type
if k == 'edge_label_index':
if edge_dir == 'out':
data[res_edge_type]['edge_label_index'] = \
torch.stack((v[1], v[0]), dim=0)
else:
data[res_edge_type]['edge_label_index'] = v
elif k == 'edge_label':
data[res_edge_type]['edge_label'] = v
elif k == 'src_index':
data[input_type[0]]['src_index'] = v
elif k in ['dst_pos_index', 'dst_neg_index']:
data[input_type[-1]][k] = v
else:
data[res_edge_type]['edge_label_index'] = v
elif k == 'edge_label':
data[res_edge_type]['edge_label'] = v
elif k == 'src_index':
data[input_type[0]]['src_index'] = v
elif k in ['dst_pos_index', 'dst_neg_index']:
data[input_type[-1]][k] = v
data[k] = v
else:
data[k] = v
elif hetero_sampler_out.metadata is not None:
Expand Down
Loading