Training on obg-molhiv dataset

[BalaDhinesh]

Thanks a lot for posting the code! I'm quite new to GNN, and I am facing errors in training the GIN model with degree quant support with the OGB-MOLHIV binary classification dataset.
However, I face an issue in the below line:

degree-quant/dq/quantization.py

Line 57 in 257d6bc

output.mul_(inv_scale).add_(zero_point)

The error is:

RuntimeError: result type Float can't be cast to the desired output type Long

The inv_scale and zero_point values I obtained are 15.9375 and -128.0, respectively.

I have only changed the dataset inside the get_dataset() function. Do I need to consider anything else? I fixed the dataset transform bug as reported by issue #1

Please help me with how to resolve this.

Thanks!!