Alex Morehead

I second @donglg1309's question. It looks like they haven't added either the source code or the data to train or test RFDiffusion yet.

Agreed. A PyTorch/Lighting 2.0 version of `openfold` would be very welcome.

Hi, @mec066. Thanks for reaching out about our work. To start with, which version of TensorFlow/TensorFlow Lite have you installed? And second, which version of the Raspberry Pi (and its...

Hi, @javimarsal. I have not personally tested our pipeline with newer versions of `scipy`, however, I believe it should be safe for you to install the closest (newer) version of...

Agreed, @pengzhangzhi. A PDF copy would be nice. For locally converting markdown to a PDF, `pandoc` seems pretty well known. However, in terms of GitHub automation for this, solutions seem...

@carmocca, great point. I'll open up an issue for PyTorch as well, linked to this one for Lightning. However, for the time being (since it may take a while for...

@awaelchli, yes, the stack trace is as follows. ```bash Traceback (most recent call last): File "", line 1, in File "/home/acmwhb/mambaforge/envs/GCPNet/lib/python3.10/site-packages/lightning/pytorch/core/module.py", line 1552, in load_from_checkpoint loaded = _load_from_checkpoint( File "/home/acmwhb/mambaforge/envs/GCPNet/lib/python3.10/site-packages/lightning/pytorch/core/saving.py",...

@awaelchli, You have described it perfectly. The checkpoints I am trying to load on a CPU-only machine contain full TorchMetrics objects in them unintentionally. Seems this is not best practice...

Hey, @gil2rok. Yes, Lightning's `strategy` class is what I had in mind here. I wanted to make it possible with this template to use any arbitrary (e.g., advanced) training strategy...

@dhibbit, did you ever find a way to suppress warnings or logs coming from calls to the `rdDetermineBonds.DetermineBonds` function? I also tried calling `RDLogger.DisableLog("rdApp.*")` to no avail. @greglandrum, any ideas...