Jin Su

> Hello, extensive use of Colab's GPU requires some tricks and additional funding. I primarily conduct wet lab experiments and am not very strong in machine learning. Could you please...

Hi, The prompt means that you have initialized a classifier behind the SaProt backbone. If you want to use it to make predictions, you have to fine-tune the classifier on...

Hi, thank you for testing it out! Could you give an example to reproduce the error? Additionally, you could submit a PR for the modified file so I can check...

As to my knowledge, protein structure predicted by AlphaFold2 only contains single chain. So if your PDB file contains multiple chains, probably it is not predicted by AF2 and therefore...

Hello, you can refer to this issue https://github.com/westlake-repl/SaProt/issues/17 for more details!

Hi, Thank you for your interest in our work! Could you explain more about how you define "distinct class"? The EC and GO tasks are multiple binary classification tasks, which...

Hi, We have thought about it before but we thought it would make the relation among files to be more complicated and may cause some unexpected errors when your execution...

In real situation it hardly happens that two ``utils`` directories conflict as every file in SaProt will load specific functions defined in ``utils``, e.g. ``utils.metrics import count_f1_max``. This would lead...

We have already uploaded a ``__init__.py`` for outer import. You can clone the repo again and try it out:)

Hi! You can definitely use the config to train a smaller SaProt. The line 14 doesn't mean the model is trained. It is just the name of the training log....