Ziyao Li

The code was tested on CUDA 11.0 (the default setting), and we reckoned in previous experiments that slight differences happened across different CUDA versions.

Hi @rabbit-0001 is this issue solved?

> @ZiyaoLi does that mean `convert_unifold_to_alphafold.py` was created somewhere? :D well this actually means that we do not plan to work on this feature, because of Guolin's comment https://github.com/dptech-corp/Uni-Fold/issues/14#issuecomment-1207314159 ,...

The inference code is released. One can try the [Colab notebook](https://colab.research.google.com/github/dptech-corp/Uni-Fold/blob/main/notebooks/unifold.ipynb).

Currently we do not release the scripts, because it has heavy dependency on how you organize the label data. We are soon releasing the entire PDB training set, perhaps in...

Can you please provide more info on the code you are using and the failure message? The code is expected to solve multi-node training with torch.distributional, so to me it...

I'm in the same position and I tried to add this `--tmscore-threshold` argument. However no matter how much the threshold is used (even with 0.999) the clustering algorithm cannot distinguish...

I also tried to print the tmscores calculated and the results seem to be correct. The following results are made with `foldseek easy-search data/ data/ aln tmp --format-output query,target,alntmscore,prob` ....

> Thank you for reporting this. This was a bug in the `structurerescorediagonal` code. It did not respect the tmscore threshold properly. I fixed it now and the following command...

Thanks for the feedback. From the given info I guess you'll have to install `wandb` and upgrade your `pytorch` to 2.0.0 before installing unicore. We'll look into the unicore repo...