search

zhangruochi / pepland

Apache License 2.0
13 stars 2 forks source link

Example data for training #1

Open amin-sagar opened 9 months ago

amin-sagar commented 9 months ago

Hello. Thanks for this absolutely fantastic work. Would it be possible to share an example file for training/test/validation dataset? It doesn't have to be real data but just to understand the format. Is it just a CSV with smiles and the value to train on? Or are there any other requirements? Best, Amin.

zhangruochi commented 9 months ago

Hello. Thanks for this absolutely fantastic work. Would it be possible to share an example file for training/test/validation dataset? It doesn't have to be real data but just to understand the format. Is it just a CSV with smiles and the value to train on? Or are there any other requirements? Best, Amin.

Hello Amin,

Thank you very much for your kind words and interest in our work. We're currently in the process of preparing our formal paper for submission, and we're also organizing the relevant data for sharing as soon as they're ready.

Best regards, Richard.

zhangruochi commented 9 months ago

Hello. Thanks for this absolutely fantastic work. Would it be possible to share an example file for training/test/validation dataset? It doesn't have to be real data but just to understand the format. Is it just a CSV with smiles and the value to train on? Or are there any other requirements? Best, Amin.

Hi Amin, I have added some example data for two step training. You can see this in the /data directory.

Cheers, Richard.

amin-sagar commented 9 months ago

Thanks @zhangruochi The uploaded csv files seems to have only smiles with no values. Does this mean that I just need to have another column with the values like solubility etc. and it should work? Best, Amin.

zhangruochi commented 9 months ago

uploaded csv

Hi Admin,

The updated csv files belong to the two-stage pre-training data. Regarding downstream tasks such as cell penetration ability prediction and binding affinity prediction, I still need some time to organize. I apologize for the inconvenience.

amin-sagar commented 9 months ago

Thanks @zhangruochi Apologies for being too excited to try this on my data :smiley:. All the best for the paper submission.