Add GCN-BMP by mughetto · Pull Request #71 · AstraZeneca/chemicalx

mughetto · 2022-02-02T09:12:34Z

Summary

First draft of implementation for GCN-BMP. Closes Issue #21 .

Unit tests provided for these changes
Documentation and docstrings added for these changes

Changes

Add GCN-BMP model. Some parts might be moved in other files/projects:
- Circular correlation function.
- Highway update
- Attention-based pooling/readout
Add example for the model.

Caveats

The number of relationships (for the Relational Conv layers) is configured in the model initialization. It should be done via some dataset-level constant.

Differences with the original paper

In the original GCN-BMP the feature vectors are initialized randomly based on the atomic number of the node. I couldn't find easily the implementation in their code so backed up on the features already provided in the PairedBatches.
Rel Conv layers are followed by a sigmoid activation before the Highway Update. This should only affect the stability.
Highway Update used is the one proposed in the original Highway network paper. Best for reusability imho and shouldn't affect the gating behavour.

Performance

Example's ROC AUC: 0.749128

…douts)

benedekrozemberczki

Generally looks good, typing is missing in certain places. The latest version of black should be used for linting. The auxiliary layers should be in the readme on some name space - also you should consider adding them to the README.

chemicalx/models/gcnbmp.py

examples/gcnbmp_examples.py

- Chunked iteration of the BMP backbone for better readability

chemicalx/models/gcnbmp.py

benedekrozemberczki · 2022-02-02T15:09:24Z

@cthoyt I overtook this.

cthoyt · 2022-02-02T15:30:46Z

@benedekrozemberczki let me know when you're done and I'll take a final pass

benedekrozemberczki · 2022-02-02T16:00:04Z

@cthoyt Ready for review.

codecov-commenter · 2022-02-02T16:13:53Z

Codecov Report

Merging #71 (42eadbf) into main (2fb3421) will increase coverage by 0.35%.
The diff coverage is 97.87%.

@@            Coverage Diff             @@
##             main      #71      +/-   ##
==========================================
+ Coverage   94.01%   94.36%   +0.35%     
==========================================
  Files          29       29              
  Lines         869      959      +90     
==========================================
+ Hits          817      905      +88     
- Misses         52       54       +2

Impacted Files	Coverage Δ
chemicalx/models/gcnbmp.py	`97.61% <97.59%> (-2.39%)`	⬇️
tests/unit/test_models.py	`100.00% <100.00%> (ø)`

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 2fb3421...42eadbf. Read the comment docs.

* linting * GCNBMP Scatter Reduction fix * Using Rel Conv Layers instead of RGCN Model (avoid unecessary sum readouts) * Added docstrings and fixed highway update implementation * Make number of relationship configurable * little help of black for linting * Cleaning upuseless imports * Sharing attention between right and left side * Adding reference to GCNBMP docstring * Type hinting everything * Fixing docstring in example * - Removing type hints in docstrings as they were added to signatures - Chunked iteration of the BMP backbone for better readability * Ading more-itertools as a dependecy * Using pairwise for encoder construction * Adding missing docstrings * Fixing linting and precommit hook * Fixing the citation back to what is in main * Tests,formatting,example * Tests,formatting,example * GCNBMP * Cleanup Co-authored-by: kcvc236 <kcvc236@seskscpg057.prim.scp> Co-authored-by: Rozemberczki <kmdb028@astrazeneca.net> Co-authored-by: kcvc236 <kcvc236@seskscpg059.prim.scp> Co-authored-by: Charles Tapley Hoyt <cthoyt@gmail.com>

* CASTER layer implementation - only supervised training stage - input dimensionality assumed to be correct * Apply black and reorganize * Move loss into its own module * Update caster.py * Reduce diff on citation * Implement DeepDrug model (#68) * WIP: model forward pass works, not tested * added dropout and batch norm * WIP: made DeepDrug example, not tested * moved to using layers only, not GCN torchdrug model * docstring * added dropout and made context feats optional * added DeepDrug unit test * deepdrug self attribute fix * docstring update * unpack method update (when no context feats used) * isort * fixed test setting (context_channels) * fixed testing without context * black * RST fix * RST fix * more pythonic loop + swap i to _ * removed context feat support in DeepDrug * removed context handling from testing DeepDrug * fixed examples DeepDrug, no context handling, decreased epochs 100->20 * removed unused import * used a wrapper for calling the same layers on pairs of batches * used a wrapper for calling the same layers on pairs of batches * docstring fix * Abstract process applied to left and right sides * Apply black * Cleanup Co-authored-by: Charles Tapley Hoyt <cthoyt@gmail.com> * Add GCN-BMP (#71) * linting * GCNBMP Scatter Reduction fix * Using Rel Conv Layers instead of RGCN Model (avoid unecessary sum readouts) * Added docstrings and fixed highway update implementation * Make number of relationship configurable * little help of black for linting * Cleaning upuseless imports * Sharing attention between right and left side * Adding reference to GCNBMP docstring * Type hinting everything * Fixing docstring in example * - Removing type hints in docstrings as they were added to signatures - Chunked iteration of the BMP backbone for better readability * Ading more-itertools as a dependecy * Using pairwise for encoder construction * Adding missing docstrings * Fixing linting and precommit hook * Fixing the citation back to what is in main * Tests,formatting,example * Tests,formatting,example * GCNBMP * Cleanup Co-authored-by: kcvc236 <kcvc236@seskscpg057.prim.scp> Co-authored-by: Rozemberczki <kmdb028@astrazeneca.net> Co-authored-by: kcvc236 <kcvc236@seskscpg059.prim.scp> Co-authored-by: Charles Tapley Hoyt <cthoyt@gmail.com> * Implement DeepDDI model (#63) * update: Add deepddi model * update: Add deepddi examples * update: Add deepddi test case * Style: deepddi model * Style: deepddi model * Style: deepddi_examples.py * Update deepddi.py * Update deepddi.py Co-authored-by: Charles Tapley Hoyt <cthoyt@gmail.com> * CASTER review fixes * flake8 fixes * CASTER: typing fix Co-authored-by: Andriy Nikolov <kgsq682@astrazeneca.net> Co-authored-by: Charles Tapley Hoyt <cthoyt@gmail.com> Co-authored-by: Piotr Grabowski <3966940+kajocina@users.noreply.github.com> Co-authored-by: Michaël Ughetto <michael.ughetto@astrazeneca.com> Co-authored-by: kcvc236 <kcvc236@seskscpg057.prim.scp> Co-authored-by: Rozemberczki <kmdb028@astrazeneca.net> Co-authored-by: kcvc236 <kcvc236@seskscpg059.prim.scp> Co-authored-by: walter <32014404+hzcheney@users.noreply.github.com>

kcvc236 and others added 7 commits January 31, 2022 13:34

linting

53b902d

GCNBMP Scatter Reduction fix

0a428e6

Using Rel Conv Layers instead of RGCN Model (avoid unecessary sum rea…

7f3c3d9

…douts)

Added docstrings and fixed highway update implementation

ef5f8bb

Make number of relationship configurable

088a553

little help of black for linting

6ba57e9

Cleaning upuseless imports

04f23ac

mughetto requested a review from benedekrozemberczki February 2, 2022 09:12

mughetto self-assigned this Feb 2, 2022

benedekrozemberczki suggested changes Feb 2, 2022

View reviewed changes

Sharing attention between right and left side

77924db

cthoyt reviewed Feb 2, 2022

View reviewed changes

chemicalx/models/gcnbmp.py Outdated Show resolved Hide resolved

cthoyt reviewed Feb 2, 2022

View reviewed changes

chemicalx/models/gcnbmp.py Outdated Show resolved Hide resolved

Adding reference to GCNBMP docstring

39d1d30

cthoyt added the model label Feb 2, 2022

Type hinting everything

455fb2c

cthoyt reviewed Feb 2, 2022

View reviewed changes

chemicalx/models/gcnbmp.py Outdated Show resolved Hide resolved

cthoyt reviewed Feb 2, 2022

View reviewed changes

chemicalx/models/gcnbmp.py Outdated Show resolved Hide resolved

cthoyt reviewed Feb 2, 2022

View reviewed changes

examples/gcnbmp_examples.py Outdated Show resolved Hide resolved

Michaël Ughetto added 4 commits February 2, 2022 10:36

Fixing docstring in example

0fe8dcf

- Removing type hints in docstrings as they were added to signatures

549696a

- Chunked iteration of the BMP backbone for better readability

Ading more-itertools as a dependecy

da9b426

Using pairwise for encoder construction

f5ad344

benedekrozemberczki linked an issue Feb 2, 2022 that may be closed by this pull request

Add the GCNBMP model #21

Closed

Michaël Ughetto added 3 commits February 2, 2022 12:04

Adding missing docstrings

71fdd72

Fixing linting and precommit hook

20a6cca

Merge branch 'main' into 21-add-gcn-bmp

5cc5288

cthoyt reviewed Feb 2, 2022

View reviewed changes

chemicalx/models/gcnbmp.py Show resolved Hide resolved

Fixing the citation back to what is in main

03a1d20

cthoyt reviewed Feb 2, 2022

View reviewed changes

chemicalx/models/gcnbmp.py Outdated Show resolved Hide resolved

cthoyt assigned benedekrozemberczki Feb 2, 2022

benedekrozemberczki added 2 commits February 2, 2022 15:28

Tests,formatting,example

aebc003

Tests,formatting,example

68a87ed

benedekrozemberczki added 2 commits February 2, 2022 15:35

GCNBMP

e461fbf

Add DeepDrug

1361c81

benedekrozemberczki marked this pull request as ready for review February 2, 2022 15:59

Cleanup

42eadbf

benedekrozemberczki approved these changes Feb 2, 2022

View reviewed changes

benedekrozemberczki merged commit 125b6fb into main Feb 2, 2022

benedekrozemberczki deleted the 21-add-gcn-bmp branch February 2, 2022 16:16

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Add GCN-BMP#71

Add GCN-BMP#71
benedekrozemberczki merged 23 commits intomainfrom
21-add-gcn-bmp

mughetto commented Feb 2, 2022

Uh oh!

benedekrozemberczki left a comment •

edited

Loading

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

benedekrozemberczki commented Feb 2, 2022

Uh oh!

cthoyt commented Feb 2, 2022

Uh oh!

benedekrozemberczki commented Feb 2, 2022

Uh oh!

codecov-commenter commented Feb 2, 2022

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

Conversation

mughetto commented Feb 2, 2022

Summary

Changes

Caveats

Differences with the original paper

Performance

Uh oh!

benedekrozemberczki left a comment • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Choose a reason for hiding this comment

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

Uh oh!

benedekrozemberczki commented Feb 2, 2022

Uh oh!

cthoyt commented Feb 2, 2022

Uh oh!

benedekrozemberczki commented Feb 2, 2022

Uh oh!

codecov-commenter commented Feb 2, 2022

Codecov Report

Uh oh!

Reviewers

Assignees

Labels

Projects

Milestone

Development

Uh oh!

4 participants

benedekrozemberczki left a comment •

edited

Loading