Validation on different dataset

[P260]

Hi,
I am using your model to use pretrained model and fine-tuning on my own dataset. Is it possible to validate my fine-tuned model on another different validation dataset? I want to keep it completely separate. Is this possible with pretrain-gnns chem model? If yes, how can I approach?

Thanks.

@weihua916

[weihua916]

Yes, I think the most relevant code is this: https://github.com/snap-stanford/pretrain-gnns/blob/master/chem/finetune.py

It reads the pre-trained model, finetunes it, and evaluates it.

[P260]

Hi @weihua916 , yes I understood that. My concern is I want to check 100% as validation set, instead of splitting in train, test, valid. Is that possible with finetune.py script?

[weihua916]

You can just skip the training part.