Adds Documentation and minor quality of life fixes #439

jnwei · 2024-05-13T08:33:20Z

The main contribution of this PR is the Documentation, which will be hosted on readthedocs.io

A few quality of life changes are also included:

Adds scripts for creating the OpenFold training set from the datasets that are stored on RODA. We will aim to host the processed datasets on RODA as well.
Adds a script for converting OpenFold v1 weights into OpenFold v2 weights; the script renames some of the modules in OpenFold.
Adds --experiment_config_json option to more easily edit model config settings in openfold/config.py
Other minor bugfixes

- much faster due to the use of threading and mp - also supports sharding

…ugfix

- much faster due to the use of threading and mp - also supports sharding

…openfold into setup-improvements

…ights only, without parsing a time step.

…a json.

This makes it more straightforward to create an alignment database directly from the flattened RODA downloads

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.

…a environment bin

- much faster due to the use of threading and mp - also supports sharding

…xtra subtests

Fixed a small typo in permutation unit test docstring

…al-scripts Duplicate expansion support

…ork better

…ecified.

christinaflo and others added 30 commits September 8, 2023 13:28

Scripts from Lukas to be used in improved setup process

afd9198

Add improved alignment-db creation script

a3bb3c4

- much faster due to the use of threading and mp - also supports sharding

updates zero_to_fp32.py for new deepspeed version and import_weight b…

1df591b

…ugfix

Adjust weight conversion and add a script for converting checkpoints.

260592e

bugfixes and adds a section to convert optim files

775f77d

Add log statement to weight conversion script

e31e037

changes reloading weights to account for possible state_dict headers.

ddfccd5

Scripts from Lukas to be used in improved setup process

50949b9

Add improved alignment-db creation script

7091820

- much faster due to the use of threading and mp - also supports sharding

Merge branch 'setup-improvements' of https://github.com/aqlaboratory/…

1195299

…openfold into setup-improvements

Updates commit value for OpenFold notebook.

9f346d3

Merge weight-loading changes into setup-improvements

9776b69

removes outputs from OpenFold.ipynb

f1cd138

Changes resume_model_weights_only flag in train_openfold to reload we…

f4df721

…ights only, without parsing a time step.

Adds experiment_config_json for setting custom configurations with …

d1fe024

…a json.

adds config recording to wandb

6ba0a59

Improve type hints and formatting

77860bb

Add default shard number

e678050

Add duplicate chain file support to alignment DB script

ee0c5db

This makes it more straightforward to create an alignment database directly from the flattened RODA downloads

Initial commit for sphinx documentation.

e41e651

update license file

fdd4e1d

DNS: Local profiler changes

d841829

changes required for pytorch2

ce000c6

psivant local pl2 upgrades without mpi configuration

bbf989a

clear up some space for docker ci

81c847b

run_pretrained_openfold.py: Change default paths of binary to be cond…

cb3ff41

…a environment bin

fix path error in scripts/utils.py and logging fix

09e74c7

Scripts from Lukas to be used in improved setup process

76b1ec9

Add improved alignment-db creation script

eb4d1e5

- much faster due to the use of threading and mp - also supports sharding

jnwei and others added 28 commits May 9, 2024 17:18

Adds mkl version to environment.yml

60eff63

make space for docker CI

cf367df

Shorten README.md main page.

7f8d124

adds mmseqs2 to environment.yml for clustering

9a6deab

Update training OpenFold docs with correct paths.

11d5fdf

Adds example directory

d61f585

make sure padded asym_id won't affect permutation steps

17b8c14

fixed bugs in unittests for multi-chain permutation. now working on e…

0df04f3

…xtra subtests

remove unnecessary lines

d1a32aa

restore to the verison on main

77cb413

added typing hints and fixed some comments

f7571e2

make sure no padded features are going to be selected as anchors

6d41838

fixed typing errors; added more comments

04d6378

added comments

bc24032

update comments;fixed typos

9a6eb64

Update tests and comments

6f1329e

fixed typing error of anchor_gt_residue

d968098

Update test_permutation.py

32765d2

Fixed a small typo in permutation unit test docstring

Merge branch 'docs' into setup-improvements

f561cec

edits to inference documentation and script for parameters

fcb7796

fix typo in environment.yml

6cba403

Merge pull request #419 from aqlaboratory/setup-improvements_addition…

29b5823

…al-scripts Duplicate expansion support

Merge branch 'main' into setup-improvements

6706864

small fix to docker-image ci workflow

49804a5

Removes mpi4py from environment.yml -- not supported for Docker builds

6384d79

updates biopython version to 1.80 in colab notebook

ffc9b3f

Updates biopython version to 1.83 in colab notebook, which seems to w…

97dae6c

…ork better

in scripts/utils.py account for case where no conda environment is sp…

d8117ce

…ecified.

jnwei merged commit f434a27 into main May 13, 2024
3 checks passed

jnwei deleted the setup-improvements branch May 13, 2024 10:02

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Adds Documentation and minor quality of life fixes #439

Adds Documentation and minor quality of life fixes #439

jnwei commented May 13, 2024

Adds Documentation and minor quality of life fixes #439

Adds Documentation and minor quality of life fixes #439

Conversation

jnwei commented May 13, 2024