separate docking scorers into separate files#57
Merged
Conversation
…e without depending on the others
mariacm12
approved these changes
Jul 2, 2025
apayne97
changed the title
ijpulidos
added a commit
that referenced
this pull request
Jul 15, 2025
* change install pin * normalise imports * finalise imports * try bulk replace imports * Revert "finalise imports" This reverts commit baad442. * Revert "Revert "finalise imports"" This reverts commit 840abd5. * Revert "try bulk replace imports" This reverts commit 1cb8a82. * bulk replace again * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add a shortcut for not training if all epochs have already been trained for. * Add weight decay handling in Trainer. * Fix weight_decay math. * Use torch.no_grad instead of model.eval. * Add model.eval() and model.train() calls. * Add Trainer weight decay loss. * Switch default batch size to be -1. * Add device Field for DatasetConfig to automatically move data to given device. * Add option for batch gradient normalization. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix comment. * Add forgotten optimizer step. * Remove print statement. * bump CI * Add classes for L1Loss and SmoothL1Loss. * Fix loss calculation for multiple prediction/loss values. * Add LossFunctionConfig options for new loss classes. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add CLI arg for LayerNorm option. * Fix typo. * Handle missing column. * Some plotting improvements. * Add convenience methods for loading in and formatting groups of TrainingPredictionTrackers. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add multiprocessing capability to TPT loading functions. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add handling for no extract_epochs passed. * Add a shortcut for not training if all epochs have already been trained for. * Add weight decay handling in Trainer. * Fix weight_decay math. * Use torch.no_grad instead of model.eval. * Add model.eval() and model.train() calls. * Add Trainer weight decay loss. * Switch default batch size to be -1. * Add device Field for DatasetConfig to automatically move data to given device. * Add option for batch gradient normalization. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix comment. * Add forgotten optimizer step. * Remove print statement. * Add classes for L1Loss and SmoothL1Loss. * Fix loss calculation for multiple prediction/loss values. * Add LossFunctionConfig options for new loss classes. * Add CLI arg for LayerNorm option. * Fix typo. * Handle missing column. * Some plotting improvements. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add convenience methods for loading in and formatting groups of TrainingPredictionTrackers. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add multiprocessing capability to TPT loading functions. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add handling for no extract_epochs passed. * Fix minor errors. * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Update asapdiscovery-ubuntu-latest.yml * Update asapdiscovery-macOS-latest.yml * Update asapdiscovery-macOS-12.yml * Separate environment files (#9) * add initial ubuntu and mac latest environments and update data.yaml github workflow to point to it * add a python script to find imports in files * add plip dependency * temporarily fix failing openfe warning catching * update detect_imports script and add modeling yaml files * update data and modeling workflows * update docking CI and add docking yaml * add dataviz to docking dependencies * add simulation yamls * add simulation yamls * add modeling to simulation dependencies * add modeling to simulation dependencies * add macos to data workflow * update workflows * remove unnecessary packages from modeling workflow * add simulation workflow package * update simulation deps * add ml env and update env names * add torchdata to docking deps * add openmmtools to deps * add ml to simulation deps * remove multiprocessing from ML deps * remove multiprocessing from simulation deps * update ml envs * make mtenn simulation dep * add modeling to ml deps * add semver to docking deps * move cli_args to have lazy imports and add spectrum env files * also run on the pydantic_2 branch * bump ci * test running with multiple env files * move openeye testing * use different micromamba command * fix missing asapdiscovery-modeling install * Update ML workflow file to include MacOS and latest micromamba version. * add workflows * Rename macOS conda-envs directory. * add workflows * Environment file for cli ci ubuntu * Environment file for cli ci macos * workflow file for CLI CI * add workflows * fix workflows for simulation and spectrum * Remove MacOS from ML workflow. * add workflows * install environments for other packages * install environments for other packages * adding data missing dependencies * environment file for alchemy ubuntu-latest * environment file for alchemy macos-latest * workflow file update for alchemy * macos uncapitalized in workflows * Add pydantic-settings. * add dataviz env files * Add mamba installing dependencies in ML workflow file. --------- Co-authored-by: Mike Henry <11765982+mikemhenry@users.noreply.github.com> Co-authored-by: kaminow <kaminow@cshl.edu> Co-authored-by: Iván Pulido <2949729+ijpulidos@users.noreply.github.com> Co-authored-by: Maria Castellanos <castelm3@mskcc.org> Co-authored-by: kaminow <benjamin.kaminow@choderalab.org> * make a real workflows env file * remove workflow deps yaml * don't run workflows CI on macos * use bioconda for logomaker * fix dataviz workflow * fix dataviz workflow * update dataviz workflow install * rename dataviz env * fix dataviz env * remove modeling as a dep for data * move trim_small_chains to backend * move trim_small_chains to backend * adapt to str enum update * fix .schema import * fix .schema import * remove macos-ci workflow * fix from .backend import * move prepped structure and docking stuff to modeling and docking envs * move plip to dataviz * fix PreppedComplex use in conftests * most postera api to workflows * most postera api to workflows * fix imports * fix imports * test fintscore in dataviz * remove other subpackages for data ci workflow * fix prep tests * fix prep tests * fix imports * fix imports * fix imports * fix imports * fix imports * remove cudatoolkit dep from simulation.yaml (#39) * switch to using the new buckets (#41) * switch to using the new buckets * missed a few urls --------- Co-authored-by: Alexander Matthew Payne <31800806+apayne97@users.noreply.github.com> * Spectrum struct changes (#10) * add rmsd-coloring * Fix chain visualization * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Fix pre-commit-ci * Add the option to align struct directory and pdb file * Needed changes on seq_alignment * Implement fitness-alignment to align wrt fasta * Add more options to struct-alignment * changes to fitness-alignment * Additions to align_seq_match * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * precommit ci fixes * fix seq-align test * struct-alignment tests * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * Add structure alignment tests * Fix struct-alignment cfold test * [pre-commit.ci] auto fixes from pre-commit.com hooks for more information, see https://pre-commit.ci * bump CI * bump CI * Added a check for blast.csv file * Bump CI * TODO to fix test here with "hidden" error * Checking subprocess return code, print error * added mafft dependancy * add mafft to ubuntu yaml * remove macOS skips from tests --------- Co-authored-by: Castellanos Morales <castelm3@iscb007.mskcc.org> Co-authored-by: Castellanos Morales <castelm3@iscf019.mskcc.org> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: Maria Castellanos <castelm3@mskcc.org> Co-authored-by: hmacdope <hugomacdermott@gmail.com> Co-authored-by: Iván Pulido <2949729+ijpulidos@users.noreply.github.com> * Dropping bespokefit in codebase (#22) * Dropping bespokefit in codebase * using pydantic.v1 instead * trying to fix CI issues by pinning to torchdata 0.8.0 for compatibility with dgl 2.3 * Fixing env path for alchemy CI * fixing package name for alchemiscale * fixing package name for alchemiscale on macos * pinning openfe versions * Bump CI * pinning openfe versions for linux * pinning compatibility to minor versions * dropping python pins * dropping python pins for every package --------- Co-authored-by: Mike Henry <11765982+mikemhenry@users.noreply.github.com> Co-authored-by: chrisiacovella <chris.iacovella@gmail.com> * update cli-ci (#48) * update cli-ci * change order of OE license checks * add simulation to cli deps * add dataviz and spectrum to deps * fixing pydantic issues in alchemy (#52) * FIXING ISSUES with pydantic models to handle ScopedKey (including serialization); fixed issues with passing json files to openfe --------- Co-authored-by: Alexander Matthew Payne <31800806+apayne97@users.noreply.github.com> * separate docking scorers into separate files (#57) * separate scorers into separate files to make it possible to import one without depending on the others * fix broken imports * xfail ml scorer tests * Lazy ml imports workflows (#60) * Lazy imports for ML scorer in workflows. * Adding warning message. Experimental ML scorer. * Moving warnings before imports. Since imports are expected to fail. * XFailing ML workflow tests separately for now * Correcting the multiple parameters * Changing default for ml_score option * Moving second ML import as lazy import * Missed one import in ligand transfer docking. * Just testing new default behavior. No ML score. * set ml_score cli default to false * add try except to ml cli * remove ml score from docking test input * make ml score default off in all docking workflows * update asapdiscovery.cli.cli to comment out ml cli --------- Co-authored-by: Alexander Matthew Payne <alexpayne1997@gmail.com> * xskip ML testing in CLI package (#61) * xskip ml in cli * changed to skip in parameterize * Add scoring scripts to spectrum (#55) * Add score files to spectrum * fix package error * update data imports * Adding tests for scoring * Add vina requirements * Fix vina test to remove bad requirements * Add minimization test * calculate_rmsd-docs * Move spectrum cli to workflows * Address comments * Apply suggestions from code review Co-authored-by: Iván Pulido <2949729+ijpulidos@users.noreply.github.com> * delete hcc channel * Apply suggestions from code review Co-authored-by: Iván Pulido <2949729+ijpulidos@users.noreply.github.com> * Comments on score.py * Update asapdiscovery-spectrum/asapdiscovery/spectrum/score.py Co-authored-by: Iván Pulido <2949729+ijpulidos@users.noreply.github.com> * Improve logging * Last requested changes --------- Co-authored-by: Maria Castellanos <castelm3@mskcc.org> Co-authored-by: Iván Pulido <2949729+ijpulidos@users.noreply.github.com> --------- Co-authored-by: hmacdope <hugomacdermott@gmail.com> Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com> Co-authored-by: kaminow <benjamin.kaminow@choderalab.org> Co-authored-by: kaminow <51923685+kaminow@users.noreply.github.com> Co-authored-by: Alexander Matthew Payne <31800806+apayne97@users.noreply.github.com> Co-authored-by: kaminow <kaminow@cshl.edu> Co-authored-by: Iván Pulido <2949729+ijpulidos@users.noreply.github.com> Co-authored-by: Maria Castellanos <castelm3@mskcc.org> Co-authored-by: Alex <alex.payne@choderalab.org> Co-authored-by: Maria Castellanos <79340102+mariacm12@users.noreply.github.com> Co-authored-by: Castellanos Morales <castelm3@iscb007.mskcc.org> Co-authored-by: Castellanos Morales <castelm3@iscf019.mskcc.org> Co-authored-by: chrisiacovella <chris.iacovella@gmail.com> Co-authored-by: Alexander Matthew Payne <alexpayne1997@gmail.com>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Description
This tries to solve the same problem as #54 without breaking things as much.
Todos
Notable points that this PR has either accomplished or will accomplish.
Status
Developers certificate of origin