-
Notifications
You must be signed in to change notification settings - Fork 1
Adds an exception for when foldx is not found #165
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
Merged
Merged
Conversation
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
miguelgondu
added a commit
that referenced
this pull request
Mar 12, 2024
* Allows mutations to be lowercase, and adds regression tests * Adds test to the foldx interface * Adds eager reparation when a FoldX black box is created (#85) * Adds eager reparation when a FoldX black box is created * Adds a flag for eager repairing, removes gcc from dependencies * Refactors examples, and cleans old ones (#87) * Implements dockstring as a black box function (#89) * Implements dockstring as a black box function * Adds tests for dockstring, fixes bug when converting to SELFIES * Adds citations for TDC and Dockstring * Adding continuous toy examples (#91) * Scaffolds the continuous problems * Lints the files defining the continuous objective functions * Updates requirements for setuptools * Adds a simple example of how to use the toy continuous problems * Adds toy examples, includes a slow fixture for pytest, adds it to the dev ini of tox * Adds a parametrized test for all possible toy continuous objectives * Renames the objective tests accordingly * The (first) great documentation PR (#95) * Adds documentation to most core files * Documents the rest of the core modules, starts documenting the objective repostories * Adds documentation for some objective functions * 96 add a gfp vae surrogate modeling task (#97) * added cbas module with assets and wrapper classes * added factory init * cleanup wrapper main --------- * Implements a way to embed continuous problems in higher dimensions (#98) * Adds Styblinsky-Tang objective to continuous problems (#99) * cleaned up a bit, removed global variable * Updates the data by adding the relevant assets for gfp-cbas * Adds evaluation budgets and number-of-evaluations tracking (#102) * Adds an evaluation budget that can be strictly enforced * Adds 'optional' to the docs * Hotfix: evaluation in the aloha problem didn't work for batches * Builds automatic documentation of the API (#103) * Scaffolds automatic documentation of the API * Adds links to class implementations * Removes colons from attribute, raises, returns titles in docstrings * Fixes a colon in docstrings, to follow numpy standards * Adds documentation to the core module * Adds documentation to the objective factory module * Documents the objective script * Documents most functions inside the objective repository * Documents the remaining objective functions inside the repository * Adds several packages to the mock imports of autodoc * Removes the autogenerated rst files from git tracking * 96 add a gfp vae surrogate modeling task (#105) * added cbas module with assets and wrapper classes * added factory init * cleanup wrapper main * added cbas module with assets and wrapper classes * restructure vae asset location * updated vae location * added SequenceGP model (Listgarten) and persisted assets * add Cbas GP npy files --------- * add npy assets to setup.cfg , cleanup rebase mistakes (#106) * 108 cbas available problem factory only returns gp (#109) * add problem factories, register multiple sub objectives --------- * Adds a test of `force_isolation` (#110) * Adds a test of force isolation * Updates the aloha environment to run on dev branch instead of master * Adds an environment yaml for basic chemistry objectives (#111) * Adds an environment yaml for basic objectives * Adds chemistry-related requirements * Removes biopython from the new requirements * Installs poli from the dev branch instead of master (#112) * Updates envs to install from dev instead of master * Modifies the poli installation for some environments, adds dependencies to installing * Moves required arguments to the top of objective factories (#114) * Moves required arguments to the top of objective factories * Moves the documentation of toy continuous objectives kwargs * Small improvements on chemistry and foldx black boxes (#115) * removes alphabet from chemistry boxes, defaults to 1 in foldx batch size * Changes an error message to reflect changes * added LamBO repo as dependency * ensured that base candidates are always among the first given sequences * adhering to coding standard * Adds immediate feedback on create using prints (#120) * remove CBas imports from __init__ (#123) * Allows flat inputs (i.e. `[b,]` instead of `[b, L]`) to black-boxes (#125) * Re-implements the logic of alignment in the abstract black box call * Implements an alignment input checker * Removes drd3 from TDC from the latest test * Removes lambo-related test until we can automatically install it * Adds `__repr__` and `__str__` to core objects (#127) * Implements str and repr dunder methods for black boxes and problem setups * Adds str and repr methods for the process warpper and observers * Foldx5 no longer needs a `rotabase.txt` file (#128) * Removes dependence on rotabase.txt file from foldx interface * Adds backwards compatibility, copying the rotabase file if users have it * Removes commented-out check of rotabase file * Minor fixes to the `foldx_rfp_lambo` black box. (#130) * add verbose error, if FoldX fails the Lambo foldx_manager fails silently * Modifies the x0 and the black box of foldx rfp lambo, adds a basic test of evaluation * Adds a skip if lambo is not installed * 132 problem factories fixed size initial dataset rfp and gfp (#133) * add n_starting_points to GFP problem * add n_starting_points, loading pareto front PDBs is minimum, add strict evluation option for Errors on mismatch * add GFP problem type * 132 problem factories fixed size initial dataset rfp and gfp (#135) * add n_starting_points to GFP problem * add n_starting_points, loading pareto front PDBs is minimum, add strict evluation option for Errors on mismatch * add GFP problem type * bugfix: default number of starting points is number of provided files if n_starting_points is None create fails * 136 gfp cbas factory create has no problem type (#137) * added problem_type parameter to factory create * black * Cleans the readme of drd3 docking, and fixes the env name (#140) * add unique flag: drop duplicate AA sequences, correct init seq selection (#142) * 139 create dedicated rfp foldx stability sasa and revert foldx stability sasa (#143) * add RFP specific stability_sasa objective * revert foldx_stability_and_sasa to rfp independent state, no n_starting * add rfp_stability_and_sasa to module import * Bugfix: parallelization flag in `foldx` now changes batch size accordingly. (#138) * Foldx parallelization now infers the appropriate batch size from the number of workers * Updates the readme to reflect the change in kwargs of getproblem * Removing the output from `foldx` w. a flag (#145) * Adds a verbose flag to foldx interface and black boxes * Adds a test for verbose output, and adds the verbose flag to the factories * the broadcasting behavior of numpy must have changed with a new version -- now we are getting the right predictions * Adding the relevant citations (#146) * Adds a table with the relevant references to cite per black box * Updates readme, removing unnecessary text * Updates citation * Renames the workflow for the website (#147) * Adds automatic installation of `lambo` (#134) * add verbose error, if FoldX fails the Lambo foldx_manager fails silently * Modifies the x0 and the black box of foldx rfp lambo, adds a basic test of evaluation * Adds a skip if lambo is not installed * Downloads files from LaMBO's repo automatically * Only downloads PDB folders that are not there yet * Refactors the code, implementing a function for downloading assets * merges dev * Moves the warning to the black boxes --------- Co-authored-by: Richard Michael <[email protected]> * Lints using black==24.1.1 (#153) * fix GFP problem, define singular WT start, remove redundant prob type (#152) * fix GFP problem, define singular WT start, remove redundant prob type ELBO working * add missing GFP asset, oracle predictiong * Lints using black==24.1.1 * fix GFP problem, define singular WT start, remove redundant prob type ELBO working * add missing GFP asset, oracle predictiong --------- * BUGFIX update minimum at iteration (#154) Currently min_hd is not set correctly. * 155 gfp provide problem type on factory create (#156) * GFPCBas factory set problem_type on create, default gp * linting * 158 gfpcbas elbo objective misspecified (#159) * format ifix inputs * optimize neg. ELBO by minimization --------- Co-authored-by: Richard Michael <[email protected]> * Implements a new interface for `create` (#161) * Implements a new interface for create * Updates the example in the readme * Updates the examples to the new interface * Adds a test of multiple observer registration (#162) * Updates readme with link to docs * Allows for creating black boxes without instancing problem factories (#163) * scaffolds the change that would allow us to create black boxes alone * Starts re-implementing the interface to return problems, solves tests for white noise * Modifies the external instancing of black box functions * Prototypes a new way of isolating complicated black box functions * Implements a way to isolate logic that's not dependent on abstract black boxes * Fixes instancing of drd3, at the cost of instancing dockstring. * Fixes registration s.t. functions and problems have separate entries in the config * Updates white noise to the new problem interface * Adapts foldx rfp lambo to the new interface * Modifies registration s.t. we no longer need to instantiate beforehand * Adapts foldx_sasa to the new interface * Adapts foldx stability to the new interface * Adapts foldx stability and sasa to the new interface * Adapts gfp_cbas to the new api * Adapts gfp select to the new interface * Adapts penalized logp to the new interface * Adapts rasp to the new interface * Adapts rdkit logp to the new interface * Adapts qed to the new interface * Adapts the rfp problem to the new interface * Adapts SA to the new interface * Removes the previous way of importing black boxes * Adapts SMB to the new interface * Adds toy continuous problems to the roster, tests all the black box functions * Updates tests to the new interface * Updates environments inside tox ini= * Updates the basic environment for forceful isolation * Updates examples to the new interface, fixes a bug on termination of isolated functions * Adds more docs examples to tests * Adds poli lambo env to tox * Adds a (slow) test for rasp's example * Updates readme to the new interface * Fixes a couple of bugs on negative and multiobjective black boxes (#164) * Adds an exception for when foldx is not found (#165) * hot-fixes the CI for master branch * hot-fix: removes lambo from master CI * hot-fix: since we lack the space, decided to test on dev only --------- Co-authored-by: Richard Michael <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Richard Michael <[email protected]>
miguelgondu
added a commit
that referenced
this pull request
Mar 12, 2024
* Allows mutations to be lowercase, and adds regression tests * Adds test to the foldx interface * Adds eager reparation when a FoldX black box is created (#85) * Adds eager reparation when a FoldX black box is created * Adds a flag for eager repairing, removes gcc from dependencies * Refactors examples, and cleans old ones (#87) * Implements dockstring as a black box function (#89) * Implements dockstring as a black box function * Adds tests for dockstring, fixes bug when converting to SELFIES * Adds citations for TDC and Dockstring * Adding continuous toy examples (#91) * Scaffolds the continuous problems * Lints the files defining the continuous objective functions * Updates requirements for setuptools * Adds a simple example of how to use the toy continuous problems * Adds toy examples, includes a slow fixture for pytest, adds it to the dev ini of tox * Adds a parametrized test for all possible toy continuous objectives * Renames the objective tests accordingly * The (first) great documentation PR (#95) * Adds documentation to most core files * Documents the rest of the core modules, starts documenting the objective repostories * Adds documentation for some objective functions * 96 add a gfp vae surrogate modeling task (#97) * added cbas module with assets and wrapper classes * added factory init * cleanup wrapper main --------- * Implements a way to embed continuous problems in higher dimensions (#98) * Adds Styblinsky-Tang objective to continuous problems (#99) * cleaned up a bit, removed global variable * Updates the data by adding the relevant assets for gfp-cbas * Adds evaluation budgets and number-of-evaluations tracking (#102) * Adds an evaluation budget that can be strictly enforced * Adds 'optional' to the docs * Hotfix: evaluation in the aloha problem didn't work for batches * Builds automatic documentation of the API (#103) * Scaffolds automatic documentation of the API * Adds links to class implementations * Removes colons from attribute, raises, returns titles in docstrings * Fixes a colon in docstrings, to follow numpy standards * Adds documentation to the core module * Adds documentation to the objective factory module * Documents the objective script * Documents most functions inside the objective repository * Documents the remaining objective functions inside the repository * Adds several packages to the mock imports of autodoc * Removes the autogenerated rst files from git tracking * 96 add a gfp vae surrogate modeling task (#105) * added cbas module with assets and wrapper classes * added factory init * cleanup wrapper main * added cbas module with assets and wrapper classes * restructure vae asset location * updated vae location * added SequenceGP model (Listgarten) and persisted assets * add Cbas GP npy files --------- * add npy assets to setup.cfg , cleanup rebase mistakes (#106) * 108 cbas available problem factory only returns gp (#109) * add problem factories, register multiple sub objectives --------- * Adds a test of `force_isolation` (#110) * Adds a test of force isolation * Updates the aloha environment to run on dev branch instead of master * Adds an environment yaml for basic chemistry objectives (#111) * Adds an environment yaml for basic objectives * Adds chemistry-related requirements * Removes biopython from the new requirements * Installs poli from the dev branch instead of master (#112) * Updates envs to install from dev instead of master * Modifies the poli installation for some environments, adds dependencies to installing * Moves required arguments to the top of objective factories (#114) * Moves required arguments to the top of objective factories * Moves the documentation of toy continuous objectives kwargs * Small improvements on chemistry and foldx black boxes (#115) * removes alphabet from chemistry boxes, defaults to 1 in foldx batch size * Changes an error message to reflect changes * added LamBO repo as dependency * ensured that base candidates are always among the first given sequences * adhering to coding standard * Adds immediate feedback on create using prints (#120) * remove CBas imports from __init__ (#123) * Allows flat inputs (i.e. `[b,]` instead of `[b, L]`) to black-boxes (#125) * Re-implements the logic of alignment in the abstract black box call * Implements an alignment input checker * Removes drd3 from TDC from the latest test * Removes lambo-related test until we can automatically install it * Adds `__repr__` and `__str__` to core objects (#127) * Implements str and repr dunder methods for black boxes and problem setups * Adds str and repr methods for the process warpper and observers * Foldx5 no longer needs a `rotabase.txt` file (#128) * Removes dependence on rotabase.txt file from foldx interface * Adds backwards compatibility, copying the rotabase file if users have it * Removes commented-out check of rotabase file * Minor fixes to the `foldx_rfp_lambo` black box. (#130) * add verbose error, if FoldX fails the Lambo foldx_manager fails silently * Modifies the x0 and the black box of foldx rfp lambo, adds a basic test of evaluation * Adds a skip if lambo is not installed * 132 problem factories fixed size initial dataset rfp and gfp (#133) * add n_starting_points to GFP problem * add n_starting_points, loading pareto front PDBs is minimum, add strict evluation option for Errors on mismatch * add GFP problem type * 132 problem factories fixed size initial dataset rfp and gfp (#135) * add n_starting_points to GFP problem * add n_starting_points, loading pareto front PDBs is minimum, add strict evluation option for Errors on mismatch * add GFP problem type * bugfix: default number of starting points is number of provided files if n_starting_points is None create fails * 136 gfp cbas factory create has no problem type (#137) * added problem_type parameter to factory create * black * Cleans the readme of drd3 docking, and fixes the env name (#140) * add unique flag: drop duplicate AA sequences, correct init seq selection (#142) * 139 create dedicated rfp foldx stability sasa and revert foldx stability sasa (#143) * add RFP specific stability_sasa objective * revert foldx_stability_and_sasa to rfp independent state, no n_starting * add rfp_stability_and_sasa to module import * Bugfix: parallelization flag in `foldx` now changes batch size accordingly. (#138) * Foldx parallelization now infers the appropriate batch size from the number of workers * Updates the readme to reflect the change in kwargs of getproblem * Removing the output from `foldx` w. a flag (#145) * Adds a verbose flag to foldx interface and black boxes * Adds a test for verbose output, and adds the verbose flag to the factories * the broadcasting behavior of numpy must have changed with a new version -- now we are getting the right predictions * Adding the relevant citations (#146) * Adds a table with the relevant references to cite per black box * Updates readme, removing unnecessary text * Updates citation * Renames the workflow for the website (#147) * Adds automatic installation of `lambo` (#134) * add verbose error, if FoldX fails the Lambo foldx_manager fails silently * Modifies the x0 and the black box of foldx rfp lambo, adds a basic test of evaluation * Adds a skip if lambo is not installed * Downloads files from LaMBO's repo automatically * Only downloads PDB folders that are not there yet * Refactors the code, implementing a function for downloading assets * merges dev * Moves the warning to the black boxes --------- * Lints using black==24.1.1 (#153) * fix GFP problem, define singular WT start, remove redundant prob type (#152) * fix GFP problem, define singular WT start, remove redundant prob type ELBO working * add missing GFP asset, oracle predictiong * Lints using black==24.1.1 * fix GFP problem, define singular WT start, remove redundant prob type ELBO working * add missing GFP asset, oracle predictiong --------- * BUGFIX update minimum at iteration (#154) Currently min_hd is not set correctly. * 155 gfp provide problem type on factory create (#156) * GFPCBas factory set problem_type on create, default gp * linting * 158 gfpcbas elbo objective misspecified (#159) * format ifix inputs * optimize neg. ELBO by minimization --------- * Implements a new interface for `create` (#161) * Implements a new interface for create * Updates the example in the readme * Updates the examples to the new interface * Adds a test of multiple observer registration (#162) * Updates readme with link to docs * Allows for creating black boxes without instancing problem factories (#163) * scaffolds the change that would allow us to create black boxes alone * Starts re-implementing the interface to return problems, solves tests for white noise * Modifies the external instancing of black box functions * Prototypes a new way of isolating complicated black box functions * Implements a way to isolate logic that's not dependent on abstract black boxes * Fixes instancing of drd3, at the cost of instancing dockstring. * Fixes registration s.t. functions and problems have separate entries in the config * Updates white noise to the new problem interface * Adapts foldx rfp lambo to the new interface * Modifies registration s.t. we no longer need to instantiate beforehand * Adapts foldx_sasa to the new interface * Adapts foldx stability to the new interface * Adapts foldx stability and sasa to the new interface * Adapts gfp_cbas to the new api * Adapts gfp select to the new interface * Adapts penalized logp to the new interface * Adapts rasp to the new interface * Adapts rdkit logp to the new interface * Adapts qed to the new interface * Adapts the rfp problem to the new interface * Adapts SA to the new interface * Removes the previous way of importing black boxes * Adapts SMB to the new interface * Adds toy continuous problems to the roster, tests all the black box functions * Updates tests to the new interface * Updates environments inside tox ini= * Updates the basic environment for forceful isolation * Updates examples to the new interface, fixes a bug on termination of isolated functions * Adds more docs examples to tests * Adds poli lambo env to tox * Adds a (slow) test for rasp's example * Updates readme to the new interface * Fixes a couple of bugs on negative and multiobjective black boxes (#164) * Adds an exception for when foldx is not found (#165) * hot-fixes the CI for master branch * hot-fix: removes lambo from master CI * hot-fix: since we lack the space, decided to test on dev only --------- Co-authored-by: Richard Michael <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Richard Michael <[email protected]>
miguelgondu
added a commit
that referenced
this pull request
Jun 6, 2024
* changes-of-poli-in-master * Allows mutations to be lowercase, and adds regression tests * Adds test to the foldx interface * Adds eager reparation when a FoldX black box is created (#85) * Adds eager reparation when a FoldX black box is created * Adds a flag for eager repairing, removes gcc from dependencies * Refactors examples, and cleans old ones (#87) * Implements dockstring as a black box function (#89) * Implements dockstring as a black box function * Adds tests for dockstring, fixes bug when converting to SELFIES * Adds citations for TDC and Dockstring * Adding continuous toy examples (#91) * Scaffolds the continuous problems * Lints the files defining the continuous objective functions * Updates requirements for setuptools * Adds a simple example of how to use the toy continuous problems * Adds toy examples, includes a slow fixture for pytest, adds it to the dev ini of tox * Adds a parametrized test for all possible toy continuous objectives * Renames the objective tests accordingly * The (first) great documentation PR (#95) * Adds documentation to most core files * Documents the rest of the core modules, starts documenting the objective repostories * Adds documentation for some objective functions * 96 add a gfp vae surrogate modeling task (#97) * added cbas module with assets and wrapper classes * added factory init * cleanup wrapper main --------- * Implements a way to embed continuous problems in higher dimensions (#98) * Adds Styblinsky-Tang objective to continuous problems (#99) * cleaned up a bit, removed global variable * Updates the data by adding the relevant assets for gfp-cbas * Adds evaluation budgets and number-of-evaluations tracking (#102) * Adds an evaluation budget that can be strictly enforced * Adds 'optional' to the docs * Hotfix: evaluation in the aloha problem didn't work for batches * Builds automatic documentation of the API (#103) * Scaffolds automatic documentation of the API * Adds links to class implementations * Removes colons from attribute, raises, returns titles in docstrings * Fixes a colon in docstrings, to follow numpy standards * Adds documentation to the core module * Adds documentation to the objective factory module * Documents the objective script * Documents most functions inside the objective repository * Documents the remaining objective functions inside the repository * Adds several packages to the mock imports of autodoc * Removes the autogenerated rst files from git tracking * 96 add a gfp vae surrogate modeling task (#105) * added cbas module with assets and wrapper classes * added factory init * cleanup wrapper main * added cbas module with assets and wrapper classes * restructure vae asset location * updated vae location * added SequenceGP model (Listgarten) and persisted assets * add Cbas GP npy files --------- * add npy assets to setup.cfg , cleanup rebase mistakes (#106) * 108 cbas available problem factory only returns gp (#109) * add problem factories, register multiple sub objectives --------- * Adds a test of `force_isolation` (#110) * Adds a test of force isolation * Updates the aloha environment to run on dev branch instead of master * Adds an environment yaml for basic chemistry objectives (#111) * Adds an environment yaml for basic objectives * Adds chemistry-related requirements * Removes biopython from the new requirements * Installs poli from the dev branch instead of master (#112) * Updates envs to install from dev instead of master * Modifies the poli installation for some environments, adds dependencies to installing * Moves required arguments to the top of objective factories (#114) * Moves required arguments to the top of objective factories * Moves the documentation of toy continuous objectives kwargs * Small improvements on chemistry and foldx black boxes (#115) * removes alphabet from chemistry boxes, defaults to 1 in foldx batch size * Changes an error message to reflect changes * added LamBO repo as dependency * ensured that base candidates are always among the first given sequences * adhering to coding standard * Adds immediate feedback on create using prints (#120) * remove CBas imports from __init__ (#123) * Allows flat inputs (i.e. `[b,]` instead of `[b, L]`) to black-boxes (#125) * Re-implements the logic of alignment in the abstract black box call * Implements an alignment input checker * Removes drd3 from TDC from the latest test * Removes lambo-related test until we can automatically install it * Adds `__repr__` and `__str__` to core objects (#127) * Implements str and repr dunder methods for black boxes and problem setups * Adds str and repr methods for the process warpper and observers * Foldx5 no longer needs a `rotabase.txt` file (#128) * Removes dependence on rotabase.txt file from foldx interface * Adds backwards compatibility, copying the rotabase file if users have it * Removes commented-out check of rotabase file * Minor fixes to the `foldx_rfp_lambo` black box. (#130) * add verbose error, if FoldX fails the Lambo foldx_manager fails silently * Modifies the x0 and the black box of foldx rfp lambo, adds a basic test of evaluation * Adds a skip if lambo is not installed * 132 problem factories fixed size initial dataset rfp and gfp (#133) * add n_starting_points to GFP problem * add n_starting_points, loading pareto front PDBs is minimum, add strict evluation option for Errors on mismatch * add GFP problem type * 132 problem factories fixed size initial dataset rfp and gfp (#135) * add n_starting_points to GFP problem * add n_starting_points, loading pareto front PDBs is minimum, add strict evluation option for Errors on mismatch * add GFP problem type * bugfix: default number of starting points is number of provided files if n_starting_points is None create fails * 136 gfp cbas factory create has no problem type (#137) * added problem_type parameter to factory create * black * Cleans the readme of drd3 docking, and fixes the env name (#140) * add unique flag: drop duplicate AA sequences, correct init seq selection (#142) * 139 create dedicated rfp foldx stability sasa and revert foldx stability sasa (#143) * add RFP specific stability_sasa objective * revert foldx_stability_and_sasa to rfp independent state, no n_starting * add rfp_stability_and_sasa to module import * Bugfix: parallelization flag in `foldx` now changes batch size accordingly. (#138) * Foldx parallelization now infers the appropriate batch size from the number of workers * Updates the readme to reflect the change in kwargs of getproblem * Removing the output from `foldx` w. a flag (#145) * Adds a verbose flag to foldx interface and black boxes * Adds a test for verbose output, and adds the verbose flag to the factories * the broadcasting behavior of numpy must have changed with a new version -- now we are getting the right predictions * Adding the relevant citations (#146) * Adds a table with the relevant references to cite per black box * Updates readme, removing unnecessary text * Updates citation * Renames the workflow for the website (#147) * Adds automatic installation of `lambo` (#134) * add verbose error, if FoldX fails the Lambo foldx_manager fails silently * Modifies the x0 and the black box of foldx rfp lambo, adds a basic test of evaluation * Adds a skip if lambo is not installed * Downloads files from LaMBO's repo automatically * Only downloads PDB folders that are not there yet * Refactors the code, implementing a function for downloading assets * merges dev * Moves the warning to the black boxes --------- * Lints using black==24.1.1 (#153) * fix GFP problem, define singular WT start, remove redundant prob type (#152) * fix GFP problem, define singular WT start, remove redundant prob type ELBO working * add missing GFP asset, oracle predictiong * Lints using black==24.1.1 * fix GFP problem, define singular WT start, remove redundant prob type ELBO working * add missing GFP asset, oracle predictiong --------- * BUGFIX update minimum at iteration (#154) Currently min_hd is not set correctly. * 155 gfp provide problem type on factory create (#156) * GFPCBas factory set problem_type on create, default gp * linting * 158 gfpcbas elbo objective misspecified (#159) * format ifix inputs * optimize neg. ELBO by minimization --------- * Implements a new interface for `create` (#161) * Implements a new interface for create * Updates the example in the readme * Updates the examples to the new interface * Adds a test of multiple observer registration (#162) * Updates readme with link to docs * Allows for creating black boxes without instancing problem factories (#163) * scaffolds the change that would allow us to create black boxes alone * Starts re-implementing the interface to return problems, solves tests for white noise * Modifies the external instancing of black box functions * Prototypes a new way of isolating complicated black box functions * Implements a way to isolate logic that's not dependent on abstract black boxes * Fixes instancing of drd3, at the cost of instancing dockstring. * Fixes registration s.t. functions and problems have separate entries in the config * Updates white noise to the new problem interface * Adapts foldx rfp lambo to the new interface * Modifies registration s.t. we no longer need to instantiate beforehand * Adapts foldx_sasa to the new interface * Adapts foldx stability to the new interface * Adapts foldx stability and sasa to the new interface * Adapts gfp_cbas to the new api * Adapts gfp select to the new interface * Adapts penalized logp to the new interface * Adapts rasp to the new interface * Adapts rdkit logp to the new interface * Adapts qed to the new interface * Adapts the rfp problem to the new interface * Adapts SA to the new interface * Removes the previous way of importing black boxes * Adapts SMB to the new interface * Adds toy continuous problems to the roster, tests all the black box functions * Updates tests to the new interface * Updates environments inside tox ini= * Updates the basic environment for forceful isolation * Updates examples to the new interface, fixes a bug on termination of isolated functions * Adds more docs examples to tests * Adds poli lambo env to tox * Adds a (slow) test for rasp's example * Updates readme to the new interface * Fixes a couple of bugs on negative and multiobjective black boxes (#164) * Adds an exception for when foldx is not found (#165) * hot-fixes the CI for master branch * hot-fix: removes lambo from master CI * hot-fix: since we lack the space, decided to test on dev only --------- Co-authored-by: Richard Michael <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Richard Michael <[email protected]> * Updates the SMB black box to be discrete instead of continuous (#175) * Adds the Hartmann6D objective to toy problems (#177) * Adds the Branin toy function (#178) * Adds slurm scripts to gitignore * Adding the remaining PMO tasks (#180) * Adds GSK3Beta, including tests * Implements an isolation for all TDC black boxes, adds DRD2 * Adds JNK3, lacks testing until tdc is fixed * Implements an abstract black box for TDC to reduce code duplication * Adds Celecoxib rediscovery * Adds Thiothixene rediscovery * Adds troglitazone rediscovery * Adds albuterol similarity * Adds mestranol similarity * amlodipine_mpo * Adds Fexofenadine MPO * Adds the remaining MPO tasks * Forgot about Zaleplon MPO * Adds the two Hop tasks * Adds the two isomer tasks * Adds the median tasks * Adds the Valsartan SMARTS task * Changes the environment creation for poli tdc * Updates the poli tdc env to have laxer restrictions on biopython and pandas * Polishes the docstrings for the new black boxes * Adds citing information to remaining black boxes * Polishes the update and guacamol tests * Adds benchmarks (#181) * Adds a benchmarks feature, starting with guacamol, pmo and toy functions * Adds two more benchmarks for embedded functions * Adds an x0 for all toy objectives * Adds documentation to PMO, GuacaMol and toy continuous benchmarks * Adds an example of running a benchmark * Fixes the bounds in toy continuous problems * Adds Levy and Rosenbrock test functions (#182) * Hotfix: removes shifted sphere from the 2d list * Fixes the solution length for shifted sphere * Makes new functions maximizable * Adds problem names as property of benchmarks * Fixes a bug: logp and qed are discrete objectives, adds a shortcut to black boxes (#183) * Abstracts away isolated function imports (#185) * Instances the inner function inside every black-box call for FoldX, adds utility for abstracting away isolated function instantiation * Uses the new isolation interface for all isolate-able black boxes (except RFP) * Removes unused imports * Removes unused import from TDC black box, removes inner_function attribute in foldx stability * fix RFPFoldXStabilityAndSASA create, eliminate inner_function (#189) * add negate flag to invert negative GP predictions for max objectives (#191) * Adds a warning when the user is not setting an observer properly (#192) * Adds a warning when the user is not setting an observer properly * Self is not available in 3.9 * Missed a space in the warning * Hotfix: gfp cbas should be installing from dev branch * Implements a multi-observer (#194) * Allows other values than NaN for unplayable levels in SMB (#196) --------- Co-authored-by: Richard Michael <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Simon Bartels <[email protected]> Co-authored-by: Richard Michael <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
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.
Closes #124