Training on graphs with different sizes

[risufaj]

Hi,

I'm pretty new to the field of graph neural networks and pytorch geometric, so this might be quite obvious. However, is it possible to train a model on a dataset of differently sized graphs (e.g. each graph has a different number of nodes).

Let's assume that I want to make graph level predictions, would this be possible?

Thank you in advance.

[rusty1s]

Yes, this is possible and supported out-of-the-box via our DataLoader, see here. Here is an example on how to use.

[risufaj]

Brilliant, thank you very much.