openfold
openfold copied to clipboard
aqlaboratory
Trainable, memory-efficient, and GPU-friendly PyTorch reproduction of AlphaFold 2
a less obtrusive version of the caption, since github doesnt really have captions
Such a good work!!!! I am so interested to deploy your work. Could you please list the requirements of your hardware such as: GPU: 32G V100 * 1 CPU: 12...
Hi, I would like to report some issues I encountered when trying to run multimer inference: 1. in feature_pipeline.py line 81: 'num_res = int(np_example["seq_length"][0])' was giving me error and changing...
When running the multimer branch with the inference-example command in the readme, I get the following error > File "/root/openfold/openfold/utils/import_weights.py", line 630, in import_jax_weights_ > assert len(incorrect) == 0 >...
hi, I trained openfold and use the ckpt to test sequence, but it was not correct, lots of atoms overlapped together ![Uploading 微信图片_20220525170716.jpg…]()
Environment: pytorch-lightning==1.5.10 torch==1.10.0+cu113 deepspeed==0.5.9 CUDA Version: 11.2.142 Driver Version: 460.32.03 CUDNN: 8.20.00 [torch.backends.cudnn.version(), 8200] NCCL: 2.10.3 [torch.cuda.nccl.version(), (2, 10, 3)] git commit 24118e804535515398e77010e11b1312f6ff1b62 & Date: Fri May 13 13:59:38 2022...
Traceback (most recent call last): File "run_pretrained_openfold.py", line 257, in main(args) File "run_pretrained_openfold.py", line 118, in main fasta_path=fasta_path, alignment_dir=local_alignment_dir File "/data1/openfold/openfold/data/data_pipeline.py", line 575, in process_fasta msa_features = self._process_msa_feats(alignment_dir, input_sequence, _alignment_index)...
Can you upload a DummyData of each training config, e.g. initial_training, finetuning, model_1, model_2, model_3, model_4, model_5, model_1_ptm, model_2_ptm,, model_3_ptm, model_4_ptm, model_5_ptm.
Hi, when i used openmm to relax the structure, i sometimes met a cuda error as metioned below. Could anyone give me a help? `Error loading CUDA module: CUDA_ERROR_UNSUPPORTED_PTX_VERSION (222)`
Hi :), First of all, thanks for your great work! I'm trying to train _openfold_ (with just 10 evoformer blocks) on _proteinnet_ with two _quadro rtx 8000_ GPUs. My loss...
Metadata
Owner
Metadata
Trainable, memory-efficient, and GPU-friendly PyTorch reproduction of AlphaFold 2