Jennifer Wei
d8117ce373
in scripts/utils.py account for case where no conda environment is specified.
2024-05-13 16:37:11 +07:00
Jennifer Wei
29b5823e8b
Merge pull request #419 from aqlaboratory/setup-improvements_additional-scripts
...
Duplicate expansion support
2024-05-13 14:57:36 +07:00
Lukas Jarosch
04410d5eb2
Improve import formatting
2024-05-05 23:53:31 -07:00
Lukas Jarosch
78b970653f
Set CLI description to more informative module docstring
2024-05-05 23:51:41 -07:00
Lukas Jarosch
244970bfe3
Slightly improve comment
2024-05-05 23:49:59 -07:00
Lukas Jarosch
0b5c9492ff
Give script more descriptive name
2024-05-05 23:49:22 -07:00
Lukas Jarosch
e2479cb539
Add more efficient script to generate all-seqs FASTA
...
The previous data_dir_to_fasta.py script is very slow and requires fully reparsing mmCIF files. This new script is much faster and uses the sequence information from the alignment data instead. Note that this will not include chains for which alignments could not be generated, but we can't use those during training anyways.
2024-05-05 23:48:19 -07:00
Jennifer
3a4ed7534d
Removes OF copy of zero_to_fp32.py favoring deepspeed.util version
2024-05-02 03:22:55 -04:00
Jennifer
2dd481e12d
fix path error in scripts/utils.py and logging fix
2024-05-02 02:36:13 -04:00
jnwei
cfb4c19cac
run_pretrained_openfold.py: Change default paths of binary to be conda environment bin
2024-05-02 02:36:13 -04:00
Jennifer
711ba0abe2
Add log statement to weight conversion script
2024-05-02 02:36:12 -04:00
Jennifer
dd1dd64137
bugfixes and adds a section to convert optim files
2024-05-02 02:36:12 -04:00
Jennifer
0a0dbb3960
Adjust weight conversion and add a script for converting checkpoints.
2024-05-02 02:36:12 -04:00
Jennifer
d9d4bf1427
updates zero_to_fp32.py for new deepspeed version and import_weight bugfix
2024-05-02 02:36:12 -04:00
Lukas Jarosch
eb4d1e57af
Add improved alignment-db creation script
...
- much faster due to the use of threading and mp
- also supports sharding
2024-05-02 02:36:12 -04:00
Christina Floristean
76b1ec98e2
Scripts from Lukas to be used in improved setup process
2024-05-02 02:36:12 -04:00
Sachin Kadyan
df8a002dc5
Add script to download embeddings for training SoloSeq
2024-03-26 17:36:00 -04:00
Sachin Kadyan
06eac9eaa7
Fix usage example in download_openfold_soloseq_params.sh
2024-03-26 17:35:03 -04:00
Lukas Jarosch
94819bf136
Add script for expanding the alignment dir with duplicates
...
This adds support for duplicate chain expansion for the alignment dir format. This script can be run on the flattened non-redundant RODA alignments to add explicit directories for all of the duplicate chains in the duplicate_chains file, symlinked to their representative chain alignment directory.
2024-03-19 21:04:35 -07:00
Lukas Jarosch
ee0c5dbe17
Add duplicate chain file support to alignment DB script
...
This makes it more straightforward to create an alignment database directly from the flattened RODA downloads
2024-03-19 20:48:26 -07:00
Lukas Jarosch
e6780504bb
Add default shard number
2024-03-19 15:14:30 -07:00
Lukas Jarosch
77860bb762
Improve type hints and formatting
2024-03-19 15:07:22 -07:00
jnwei
9776b69634
Merge weight-loading changes into setup-improvements
2024-02-21 14:06:18 +07:00
Lukas Jarosch
70918209d4
Add improved alignment-db creation script
...
- much faster due to the use of threading and mp
- also supports sharding
2024-02-20 12:21:32 -05:00
Christina Floristean
50949b9a58
Scripts from Lukas to be used in improved setup process
2024-02-20 12:21:32 -05:00
Jennifer
e31e03781f
Add log statement to weight conversion script
2024-02-14 00:48:00 -05:00
Jennifer
775f77dd12
bugfixes and adds a section to convert optim files
2024-02-12 05:14:53 -05:00
Jennifer
260592e066
Adjust weight conversion and add a script for converting checkpoints.
2024-02-09 04:00:07 -05:00
Jennifer
1df591b06f
updates zero_to_fp32.py for new deepspeed version and import_weight bugfix
2024-02-08 04:08:35 -05:00
Christina Floristean
1606ac0834
Merge branch 'main' into multimer
2023-12-11 10:48:16 -05:00
Christina Floristean
67a00a6cd4
Remove env restart from setup script
2023-12-08 16:50:23 -05:00
Christina Floristean
9236c1e367
Merge branch 'main' into deepspeed-evo-attention
2023-12-05 08:39:34 -05:00
jnwei
4a50c9c48d
Adds Soloseq parameter download script.
2023-11-29 18:30:19 -05:00
Christina Floristean
7fb12cf5cd
Update setup script and refactor qkv prep
2023-11-13 13:12:29 -05:00
Christina Floristean
5aa549583a
Merge branch 'main' into deepspeed-evo-attention
2023-11-08 13:32:44 -05:00
Christina Floristean
f65b75fe48
Fix for loading old OF weights into refactored model
2023-11-03 14:26:18 -04:00
Christina Floristean
5eacd8b689
Merge branch 'main' into multimer
2023-10-30 16:20:13 -04:00
Christina Floristean
15850092d3
Added multimer inference to README
2023-10-30 16:15:05 -04:00
Jennifer Wei
f7dba95f0b
Merge pull request #360 from aqlaboratory/installation-updates
...
Merging installation updates and updates
2023-10-27 15:13:12 -04:00
Jennifer Wei
e3716118cd
Adds KMP_AFFINITY to conda environment.
2023-10-27 12:04:15 -07:00
Christina Floristean
d7c115371e
Merge branch 'main' into multimer
2023-10-27 12:09:39 -04:00
Sachin Kadyan
86b990d6ed
Separate out input parsing code in EmbeddingGenerator
...
Bugfix: Corrected paths for just-in-time embedding generation
2023-10-24 12:30:25 -05:00
Sachin Kadyan
8185c30775
Just-in-time embedding generation for the SoloSeq model
2023-10-24 01:03:20 -05:00
Sachin Kadyan
92835fd5e6
More cleaning of bulk embedding generation script
2023-10-23 15:14:55 -05:00
Sachin Kadyan
0026173e23
Cleaned up precompute_embeddings.py.
2023-10-22 21:48:53 -05:00
Sachin Kadyan
bcc6d97b69
New script for generating ESM embeddings in bulk
2023-10-20 21:33:44 -05:00
jnwei
d6ae9f5894
Remove conda env config setting and update to README
2023-10-20 15:44:34 -04:00
Jennifer Wei
fcba33580e
- Moves python packages to conda installation
...
instead of pip
- Adds helper line to automatically prepend conda library to
$LD_LIBRARY_PATH
2023-10-20 06:48:05 -07:00
Jennifer Wei
705c26773d
Updating $LD_LIBRARY_PATH to include conda environment library.
2023-10-17 09:10:49 -07:00
Jennifer Wei
4fde713c05
update installation scripts.
2023-10-17 08:22:46 -07:00
