Replies: 1 comment 1 reply
-
This is great. Please feel free to add it :) |
Beta Was this translation helpful? Give feedback.
1 reply
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Uh oh!
There was an error while loading. Please reload this page.
-
I have been using
TemporalData
objects to load JODIE-formatted temporal data: https://github.com/rusty1s/pytorch_geometric/blob/4d062ae84e1c0353e7c8d15481f8b6e6d3f2ecca/torch_geometric/data/temporal.pyHowever,
TemporalData
is a slightly odd datatype, so I've written my own custom code so I can wrap it in PyTorchDataLoader
objects for parallelisation (appended below).Does it make sense for me to do a PR? Most of what the following code does can be covered by using a
TemporalDataset
class to wrap aroundTemporalData
, and adding a case in theCollater
class, but I'm not sure it belongs alongside the other cases: https://github.com/rusty1s/pytorch_geometric/blob/4d062ae84e1c0353e7c8d15481f8b6e6d3f2ecca/torch_geometric/data/dataloader.py#L13Beta Was this translation helpful? Give feedback.
All reactions