openfold
openfold copied to clipboard
aqlaboratory
Trainable, memory-efficient, and GPU-friendly PyTorch reproduction of AlphaFold 2
Hi, I`am running the script prep_mmseqs_dbs.sh. I`ve done the corrections in script changing tar2exprofiledb to tsv2exprofiledb. but the script extract the files and return the following error: > uniclust30_2018_08/uniclust30_2018_08_a3m.ffdata >...
As it stands, only the attention primitives `Attention` and `GlobalAttention` are TorchScript-ed (or, for that matter, TorchScript-able) during inference. For better runtimes and memory allocation, more of the network's modules---especially...
Our recent [OpenMM 7.6 release](https://github.com/openmm/openmm/releases/tag/7.6.0) included some namespace changes that look liked they required pinning this repo to OpenMM 7.5.1. Is it OK with you folks if we propose a...
Hi, Is openfold able to use "unified memory"? When using alphafold adding the following variables export TF_FORCE_UNIFIED_MEMORY='1' export XLA_PYTHON_CLIENT_MEM_FRACTION='4.0' export TF_FORCE_GPU_ALLOW_GROWTH=true increased the size of the sequence that can be...
Can someone help me. Why am I getting this error when I run inference： Traceback (most recent call last): File "/data/lwq/openfold-1.0.0/lib/conda/envs/openfold_venv/lib/python3.7/runpy.py", line 193, in _run_module_as_main "__main__", mod_spec) File "/data/lwq/openfold-1.0.0/lib/conda/envs/openfold_venv/lib/python3.7/runpy.py", line...
When I ran the openfold-multimer branch code to infer the multi-chain amino acid sequence, I encountered the following error. Can someone tell me the solution. (openfold_venv) [lwq@alphafold2-1 openfold-multimer]$ cd /data/lwq/openfold-multimer...
- Added a `PreembeddingEmbedder` for embedding single-sequence (NUM_RESIDUE, ...) shaped embeddings as input.
Hi. I am trying to follow the documentation to install and train the model. I have successfully installed everything and have run the following commands so far, also successfully: bash...
May I ask why this error occurs, I have confirmed that the relevant database and file paths are correct as required. However, using docker does not have this error when...
I notice that `Rigid.identity` [here ](https://github.com/aqlaboratory/openfold/blob/main/openfold/model/structure_module.py#L666) requires gradient in training stage. It will be inconsistent with the stage of inference? Is it correct?
Metadata
Owner
Metadata
Trainable, memory-efficient, and GPU-friendly PyTorch reproduction of AlphaFold 2