Hi! I appreciate your nice work for 3d molecular generation!
I'm pretty newbie in 3D molecule generation tasks.
So, I'm sorry that if my question is basic.
I'm trying to sample some 3D molecules using your model. Since I want to start from scratch, I downloaded the raw SDF files you provided, named 'sdf.tar.gz (~4GB)'.
I extracted the files using the following command: tar -xvzf sdf.tar.gz.
This gave me the processed.lmdb and processed_molid2idx.pt files from the train_drug3d.py script.
However, I'm unsure how to use the other datasets.
To my understanding, in order to generate those processed files,
I need both the mol_summary.csv and split_by_molid.pt files.
However, I couldn't find how to make these files in your code or README.
Is there something I'm missing?
Thanks in advance,
Sincerely!
Hi! I appreciate your nice work for 3d molecular generation!
I'm pretty newbie in 3D molecule generation tasks.
So, I'm sorry that if my question is basic.
I'm trying to sample some 3D molecules using your model. Since I want to start from scratch, I downloaded the raw SDF files you provided, named 'sdf.tar.gz (~4GB)'.
I extracted the files using the following command: tar -xvzf sdf.tar.gz.
This gave me the processed.lmdb and processed_molid2idx.pt files from the train_drug3d.py script.
However, I'm unsure how to use the other datasets.
To my understanding, in order to generate those processed files,
I need both the mol_summary.csv and split_by_molid.pt files.
However, I couldn't find how to make these files in your code or README.
Is there something I'm missing?
Thanks in advance,
Sincerely!