openfold
openfold copied to clipboard
aqlaboratory
Trainable, memory-efficient, and GPU-friendly PyTorch reproduction of AlphaFold 2
Hey everyone, I tested both the download via aws and via huggingface and while the former works, the latter yields always this error: torch RuntimeError: PytorchStreamReader failed reading zip archive:...
I am trying to finetune AF-multimer and I start with a mock dataset of just 3 proteins. I get the following error: torch.cuda.OutOfMemoryError: CUDA out of memory. Tried to allocate...
Default Model fails with OOM, when using long_sequence_inference=True， it breaks at template offload --------------------------------------------------------------------------- AttributeError Traceback (most recent call last) Cell In[13], line 2 1 t = time.perf_counter() ----> 2...
I have downloaded the AF2 multimer and Openfold soloseq model weights. I am unable to find the openfold multimer model weights. Have the been released? If so, where do I...
reduce_max_clusters_by_max_templates -> reduce_msa_clusters_by_max_templates
Hello, I'm working on porting open fold to a Grace Hopper machine (Arm / aarch64). When trying to install according to the instructions I get: ```shell PackagesNotFoundError: The following packages...
## Background When running inference with `run_pretrained_openfold.py` and using precomputed alignments, the `parse_fasta` function is partially extracting the FASTA tag/ID from the original ID used to generate the alignments output...
## Overview This PR introduces a fully featured Local Notebook for performing inference, obtaining metrics, ranking the best model, and generating plots in a structured and reproducible manner, particularly for...
The main issue is that the parameter name did not coincide with the code. Question: could the deepspeed extra dependency on CUTLASS be baked into the docker image? I could...
Metadata
Owner
Metadata
Trainable, memory-efficient, and GPU-friendly PyTorch reproduction of AlphaFold 2