pub_soft/openfold
3
0
Fork 0
mirror of https://github.com/aqlaboratory/openfold.git synced 2026-06-04 12:44:26 +08:00
Code Issues Packages Projects Releases Wiki Activity

Commit Graph

  • 6f10686bcd Includes button for colab link and specifies GPU type Jennifer Wei 2023-11-21 15:59:27 -05:00
  • e09dd313f2 Fixes imports to colab notebook. jnwei 2023-11-21 15:49:06 -05:00
  • cdeb8d1be9 Fix for issue ##369: Bug in multimer deterministic train filter Christina Floristean 2023-11-21 15:30:12 -05:00
  • 1271a03fd5 Use ds kernel for default long seq inference Christina Floristean 2023-11-16 00:21:01 -05:00
  • 71526dff75 Config file fix Christina Floristean 2023-11-13 16:34:55 -05:00
  • aebda3d89a Config fixes for when using ds kernel Christina Floristean 2023-11-13 15:40:07 -05:00
  • f156399972 Minor test fix Christina Floristean 2023-11-13 13:33:27 -05:00
  • 7fb12cf5cd Update setup script and refactor qkv prep Christina Floristean 2023-11-13 13:12:29 -05:00
  • d8c2a9795e Fix naming error in alignment runner, add error message for missing uniprot files. Christina Floristean 2023-11-10 14:14:32 -05:00
  • 54d414e4de Return to regular kernel usage Christina Floristean 2023-11-10 13:29:08 -05:00
  • b7f35dce65 Use EvoformerFusedAttention directly to avoid all-zero bias term in column attention Christina Floristean 2023-11-08 13:57:43 -05:00
  • 5aa549583a Merge branch 'main' into deepspeed-evo-attention Christina Floristean 2023-11-08 13:32:44 -05:00
  • 7e2b36e07f Fix for feature indexing when creating output protein object Christina Floristean 2023-11-06 16:10:14 -05:00
  • c05b46749a Add description to dummy msa feats Christina Floristean 2023-11-06 14:05:05 -05:00
  • d0f5531b0b Another fix for dummy msa feature creation Christina Floristean 2023-11-06 13:50:11 -05:00
  • e67d041759 Fix for creating msa dummy features Christina Floristean 2023-11-06 13:30:18 -05:00
  • 099769d2ec Hotfix to correct seed type in inference Christina Floristean 2023-11-06 13:11:04 -05:00
  • f65b75fe48 Fix for loading old OF weights into refactored model Christina Floristean 2023-11-03 14:26:18 -04:00
  • 5fcd6ed221 Unit test fixes for when AF2 is not installed Christina Floristean 2023-11-02 15:57:23 -04:00
  • f95d9a570a Change order of numpy and pandas dependencies to keep numpy version. jnwei 2023-11-02 14:40:56 -04:00
  • f94ee668e5 Fixes for inference alignment generation Christina Floristean 2023-11-02 13:43:48 -04:00
  • 5eacd8b689 Merge branch 'main' into multimer Christina Floristean 2023-10-30 16:20:13 -04:00
  • 15850092d3 Added multimer inference to README Christina Floristean 2023-10-30 16:15:05 -04:00
  • 2338b896c7 Merge pull request #365 from aqlaboratory/dependabot/github_actions/actions/setup-python-4 Jennifer Wei 2023-10-30 14:04:51 -04:00
  • dcd809d9c2 Bump actions/setup-python from 2 to 4 dependabot[bot] 2023-10-30 14:55:31 +00:00
  • 51472e756a Merge pull request #366 from aqlaboratory/dependabot/github_actions/actions/checkout-4 Jennifer Wei 2023-10-30 10:54:56 -04:00
  • 13728b1203 Bump actions/checkout from 2 to 4 dependabot[bot] 2023-10-27 19:25:04 +00:00
  • f7dba95f0b Merge pull request #360 from aqlaboratory/installation-updates Jennifer Wei 2023-10-27 15:13:12 -04:00
  • e3716118cd Adds KMP_AFFINITY to conda environment. Jennifer Wei 2023-10-27 12:04:15 -07:00
  • 6df89c763f Merge branch 'installation-updates' of https://github.com/aqlaboratory/openfold into installation-updates Jennifer Wei 2023-10-27 11:33:01 -07:00
  • 13f0f6fe16 Adds libgcc and pytorch::pytorch to environment.yml Jennifer Wei 2023-10-27 11:30:53 -07:00
  • d7c115371e Merge branch 'main' into multimer Christina Floristean 2023-10-27 12:09:39 -04:00
  • 9e32781fd6 README: Small improvements, info regarding templates Sachin Kadyan 2023-10-25 13:42:59 -05:00
  • f68a6c694b Update readme, also reset flash attention version to 0.2.1 in environment.yml jnwei 2023-10-25 08:51:58 -04:00
  • a5c69a79c6 Test docker image nvidia/11.7.1-cudnn8-devel-ubuntu18.04 jnwei 2023-10-24 19:55:58 -04:00
  • e2bb3c4b90 Change docker image to use cudnn devel environment jnwei 2023-10-24 19:43:58 -04:00
  • 6fe34248b2 Test nvidia image 11.4.3-cudnn8-runtime-unbuntu18.04 jnwei 2023-10-24 19:32:15 -04:00
  • 5efba4425a Fix export statement jnwei 2023-10-24 18:31:13 -04:00
  • 3817d94098 Update Dockerfile to add conda path to $LD_LIBRARY_PATH jnwei 2023-10-24 18:09:42 -04:00
  • a5a86d4323 update flash-attention-version jnwei 2023-10-24 18:07:06 -04:00
  • f06657fe8a Add git to environment.yml Jennifer Wei 2023-10-24 13:54:21 -07:00
  • 2d4fe4f414 README: Update details about SoloSeq and inference-time embedding generation. Sachin Kadyan 2023-10-24 13:07:14 -05:00
  • a90da39554 Test flash-attention v0.2.1 on docker CI jnwei 2023-10-24 13:50:32 -04:00
  • 86b990d6ed Separate out input parsing code in EmbeddingGenerator Bugfix: Corrected paths for just-in-time embedding generation Sachin Kadyan 2023-10-24 12:30:25 -05:00
  • 8185c30775 Just-in-time embedding generation for the SoloSeq model Sachin Kadyan 2023-10-24 01:03:20 -05:00
  • 4c8e37644e README: Add details about using the bulk embedding generation script Sachin Kadyan 2023-10-23 20:45:56 -05:00
  • 5f5c8f2a5b Merge remote-tracking branch 'mattwthompson/ci' into installation-updates jnwei 2023-10-23 18:27:01 -04:00
  • 7666c80272 Use conventional "short hash" style Matthew W. Thompson 2023-10-23 16:13:16 -05:00
  • 582103505d Update name to pytorch-lightning Matthew W. Thompson 2023-10-23 16:01:16 -05:00
  • 736d668741 Add Dependabot for Actions Matthew W. Thompson 2023-10-23 15:53:34 -05:00
  • 32c11376d7 Update checkout action Matthew W. Thompson 2023-10-23 15:52:39 -05:00
  • f86d42f40e Switch to Miniforge3 Matthew W. Thompson 2023-10-23 15:45:33 -05:00
  • 6bf5c8cea1 Update environment YAML Matthew W. Thompson 2023-10-23 15:37:36 -05:00
  • 92835fd5e6 More cleaning of bulk embedding generation script Sachin Kadyan 2023-10-23 15:14:55 -05:00
  • 0026173e23 Cleaned up precompute_embeddings.py. Sachin Kadyan 2023-10-22 21:48:53 -05:00
  • bcc6d97b69 New script for generating ESM embeddings in bulk Sachin Kadyan 2023-10-09 21:33:44 -05:00
  • 0c20e3c989 Update BibTex Gustaf Ahdritz 2023-10-19 18:44:01 -04:00
  • d6ae9f5894 Remove conda env config setting and update to README jnwei 2023-10-20 15:44:34 -04:00
  • b3a118fc83 Updates Dockerfile to not use OpenMM patches. Jennifer Wei 2023-10-20 09:20:35 -07:00
  • 2893fd934b Merge branch 'installation-updates' of https://github.com/aqlaboratory/openfold into installation-updates Jennifer Wei 2023-10-20 07:49:19 -07:00
  • 3e3f07c7f2 Updates installation instructions in the README. Jennifer Wei 2023-10-20 07:02:11 -07:00
  • fcba33580e - Moves python packages to conda installation instead of pip - Adds helper line to automatically prepend conda library to $LD_LIBRARY_PATH Jennifer Wei 2023-10-20 06:48:05 -07:00
  • 2300f6720d Update BibTex Gustaf Ahdritz 2023-10-19 18:44:01 -04:00
  • d77a8dabea Merge branch 'installation-updates' of https://github.com/aqlaboratory/openfold into installation-updates Jennifer 2023-10-17 17:30:21 -04:00
  • fb34a0cb62 Remove openmm patches Jennifer 2023-10-17 17:10:24 -04:00
  • 705c26773d Updating $LD_LIBRARY_PATH to include conda environment library. Jennifer Wei 2023-10-17 09:10:49 -07:00
  • 4fde713c05 update installation scripts. Jennifer Wei 2023-10-17 08:22:46 -07:00
  • 7c7dffd081 Refactoring multimer perm alignment, temporary fixes before removing recycling dimension Christina Floristean 2023-10-16 21:20:38 -04:00
  • def329b880 Removed references to jax Christina Floristean 2023-10-16 16:36:01 -04:00
  • 8470b803e3 Fix multimer sampling Christina Floristean 2023-10-16 16:22:19 -04:00
  • 148533796a Merge branch 'multimer' of https://github.com/aqlaboratory/openfold into multimer Christina Floristean 2023-10-16 15:45:09 -04:00
  • 0cf1541ce1 Refactoring multimer data pipeline and permutation alignment. Christina Floristean 2023-10-16 15:40:40 -04:00
  • 7922bd57f1 Removes conda installation from installation script. Jennifer Wei 2023-10-16 10:22:43 -07:00
  • ec75fe2245 Merge pull request #359 from dingquanyu/update-kabsch_rotation Christina Floristean 2023-10-16 11:34:39 -04:00
  • cdfb0c75d8 move the kabsch rotation step to gpu Geoffrey Yu 2023-10-12 15:12:56 +02:00
  • 6381ddd6e9 README fixes for single sequence model Sachin Kadyan 2023-10-11 15:41:15 -05:00
  • e8de822e9b Improved UX: Automatically set the single seq mode flag Sachin Kadyan 2023-10-05 13:16:22 -05:00
  • c8c1239723 Separate out the seq mode configs from vanilla OF config Sachin Kadyan 2023-10-02 20:31:57 -05:00
  • b45a91ba5c Limit the MSA distillation clusters to 1 in seq mode Sachin Kadyan 2023-10-02 19:41:56 -05:00
  • 3f592307eb Reduce redundancy in seq embedding config presets Sachin Kadyan 2023-10-02 17:49:19 -05:00
  • 6aefa986a8 Fix typos Gustaf Ahdritz 2023-09-29 00:40:52 -04:00
  • 3c240cb3f2 Updated README: Running seqemb model inference Sachin Kadyan 2023-09-27 15:50:58 -05:00
  • 28334db382 Added test for no column attention Evoformer Sachin Kadyan 2023-09-18 22:42:34 -05:00
  • a7c0d0d178 Added sequence embedding mode test for model. Sachin Kadyan 2023-09-18 22:33:18 -05:00
  • 777d738a59 Added test for PreembeddingEmbedder Sachin Kadyan 2023-09-17 23:23:05 -05:00
  • 6012b9e1c1 Default value for --use_single_seq_mode arg Sachin Kadyan 2023-09-15 01:05:53 -05:00
  • 08ef6e9fb6 Add sequence embedding mode option to .core file parser Sachin Kadyan 2023-09-15 00:16:24 -05:00
  • 395a9f1ba8 Added flag in training script for using sequence embeddings Sachin Kadyan 2023-09-14 22:05:36 -05:00
  • f85d67f4f9 Bugfix for timings.json - now store timings per tag. sachinkadyan7 2023-09-13 13:19:31 -07:00
  • 9b114f28df Renamed preembedding_embedder to input_embedder sachinkadyan7 2023-09-11 22:46:28 -07:00
  • 5a8d2b78c1 Properly reading the embedding file sachinkadyan7 2023-09-11 22:45:15 -07:00
  • 36d5708cfd Added config presets for esm1b model inference sachinkadyan7 2023-09-11 22:20:06 -07:00
  • 047e69af8d Fix for a bug in data_transforms which wouldn't allow creation of MSA mask if there is only input sequence in MSA. - Set max_msa_clusters=1 in model presets for allowing the input sequence to be a MSA cluster centre. Sachin Kadyan 2022-11-01 15:50:47 -04:00
  • 624b5aa698 Added switch for sequence embedding mode to the PDB file pipeline. - In seqemb_mode, process_pdb loads sequence embedding for the PDB's protein, and a dummy MSA Sachin Kadyan 2022-10-29 16:41:54 -04:00
  • 299629903b Added documentation for some sequence embedding model changes. Sachin Kadyan 2022-10-28 17:21:31 -04:00
  • a83c6fcc3e Added inference model preset for seqemb mode. Sachin Kadyan 2022-10-28 15:34:44 -04:00
  • 8c94482aa0 Added a separate AlignmentRunner for handling seq_emb mode. - In seq_emb mode, the AlignmentRunner works only on generating templates. Sachin Kadyan 2022-10-28 15:32:40 -04:00
  • 380947c429 Added the seq_emb features to the list of features to be processed by feature pipeline , if using seq_emb mode - In seq_emb mode, add list of seq_emb features to feature_names Sachin Kadyan 2022-10-28 15:22:03 -04:00
  • f2540236b7 Changed the seq embedding tensor passed to the data pipeline to be a seqemb feature dictionary. - _process_seqemb_features now returns a dictionary instead of a tensor. Sachin Kadyan 2022-10-21 17:51:02 -04:00
  • 43d0964536 Optimized type-changing of features from numpy to torch - Bugfix: torch throws warnings when copying a tensor via initialization - Added lambda to .clone() those tensors instead Sachin Kadyan 2022-10-19 01:27:27 -04:00