RandomLinkSplit Load_csv tutorial on hetero

[therealadarsh]

Hi,
I am trying to create a new graph dataset using hetero tutorial from load_csv. I was able to add the data and sorta follow up well until the RandomLinkSplit. I keep getting this error which basically the attribute edge_index is not present. But I can see it when I print data. Any tips on what I am missing? I have added the screenshot for your reference.

[rusty1s]

That is a bit strange. Just from looking at the images, that should definitely work. Do you have an example to reproduce on our end? Does

for edge_type in data.edge_types:
    print(edge_type)
    print(data[edge_type].edge_index)

works for you?