Shengchao Liu
Shengchao Liu
Hi @voidpunk, Yes, you are right. The BBBP dataset measures the barrier permeability (or the `penetration` as you mentioned), and it is collected from this [paper](https://pubmed.ncbi.nlm.nih.gov/22612593/). We collect the datasets...
Any updates? Same issue here. We want to reproduce the paper's results.
Thanks for the reply @vidarmehr. Just want to quickly clarify a number: 1. In `bio-decagon-targets.csv`, there are 18,690 interactions. 2. In `bio-decagon-targets-all.csv`, there are 131,034 interactions, and 112,438 of them...
@kajjana It seems to be the issue of `self.node_dim`. Originally it's set as -2, and I hack it by setting it as 0. In specific, there are two ways to...
@TheShadow29 Thanks for replying. I noticed the author has been updating this repo. One thing I didn't try before is using `cudnn.benchmark = True`, which helps a lot. Besides this...
@fanninnypeom I got the same questions, as posted in #13. Another similar issue can be found in #11 . Below is my guess: | | Paper | Implementation| | ---...
@fansariadeh There's no `model.test()` in pytorch. When you train the model, call `model.train()` will activate some layers like batch-norm, zero-out. When you test the model, call `model.eval()` will modify the...
Same issue here.
@Lassser66 @godlovexiari I found the issue, and just made a pull request. Now the results should be consistent.
Hi @RJT1990, I have the same question here. We are interested in the text prompts you extracted from MoleculeNet.