Skip to content
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

Jump to bottom

Fix DataLoader in forks #3566

Merged
merged 2 commits into from Nov 26, 2021
Merged

Fix DataLoader in forks #3566

Changes from all commits
Commits
Show all changes
2 commits
Select commit Hold shift + click to select a range
2948661
fix fork
rusty1s Nov 25, 2021
1f26065
update
rusty1s Nov 25, 2021
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Jump to
Jump to file
Failed to load files.
Diff view
Diff view
19 changes: 12 additions & 7 deletions torch_geometric/data/collate.py
View file
Open in desktop
Expand Up @@ -99,18 +99,15 @@ def collate(
if (attr in follow_batch and isinstance(slices, Tensor)
and slices.dim() == 1):
repeats = slices[1:] - slices[:-1]
arange = torch.arange(len(values), device=device)
batch = arange.repeat_interleave(repeats.to(device))
batch = repeat_interleave(repeats.tolist(), device=device)
out_store[f'{attr}_batch'] = batch

# In case the storage holds node, we add a top-level batch vector it:
if (add_batch and isinstance(stores[0], NodeStorage)
and stores[0].can_infer_num_nodes):
arange = torch.arange(len(stores), device=device)
repeats = torch.tensor([store.num_nodes for store in stores],
device=device)
out_store.batch = arange.repeat_interleave(repeats)
out_store.ptr = cumsum(repeats)
repeats = [store.num_nodes for store in stores]
out_store.batch = repeat_interleave(repeats, device=device)
out_store.ptr = cumsum(torch.tensor(repeats, device=device))

return out, slice_dict, inc_dict

Expand Down Expand Up @@ -201,6 +198,14 @@ def _collate(
###############################################################################


def repeat_interleave(
repeats: List[int],
device: Optional[torch.device] = None,
) -> Tensor:
outs = [torch.full((n, ), i, device=device) for i, n in enumerate(repeats)]
return torch.cat(outs, dim=0)


def cumsum(value: Union[Tensor, List[int]]) -> Tensor:
if not isinstance(value, Tensor):
value = torch.tensor(value)
Expand Down