About pubmed dataset result

[xh3204]

I want to reproduce the result on the PubMed dataset with a 0.00 perturbation rate. However, I can't get the same results recorded in your paper while I just modify the ptb_rate to 0 in the .sh file. Does there need another hyperparameters setting since the providing .sh file corresponds to ptb_rate = 0.15?

[ChandlerBang]

Hi,

If you set ptb_rate to 0, the loaded datasets will be using the default splits in Table 1 (20 nodes per class for training).

As shown in sample.py, you can set ptb_rate to 10 to get the result on clean dataset with 10%/10%/80% for traininig/validation/test.

Thanks.