Release v0.5.0 #1342
Merged
Release v0.5.0 #1342
Conversation
This file contains 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
* Include pymc in the documentation. * Moved Parameter description to correct place at init * Moved extra information of text upwards. * Added __all__ definition to pymc * Update pypesto/sample/pymc.py Co-authored-by: Daniel Weindl <dweindl@users.noreply.github.com> * Moved back docstring to attributes --------- Co-authored-by: Daniel Weindl <dweindl@users.noreply.github.com>
* add approx grad ipopt option --------- Co-authored-by: Simon Merkt <simon.merkt@uni-bonn.com> Co-authored-by: Paul Jonas Jost <70631928+PaulJonasJost@users.noreply.github.com>
* Removed old python versions from black
* Replaced most precommit hooks for ruff.
* Changed to ruff. Currently only checking for F rules
* Solved F errors
* Solved I errors
* Solved D413 error
* Solved D errors. (Ignoring for example python files) with correct line length
* Included flake8-bandit
* Included flake8-bugbear
* Included flake8-comprehensions
* Included juypternotebooks, added pycodestyyle. Removed flake8 nd nbqa from pre-commits
* small formatting
* correct tmpfile handling
* Add D413 (blank line after last section) to ignore
* Removed non existing file from excludes
* Removed D107 from file specific, as it is in all files ignored
* Removed black and isort from pyproject.toml
* Revert D413
* removed flake8 and project from tox.removed .flake8
* fixed {} to dict()
* Added two missing notebooks to test.
* Removed depracated force_compile
* corrected workflow_comparison
* removed manual profiling from running and only as comment
* removed outputs
* removed workflow_comparison.ipynb
---------
Co-authored-by: Daniel Weindl <daniel.weindl@helmholtz-munich.de>
* fix calc return if simulation failed * Daniel review change * Fix the extension
* PetabImporter checks for existing amici model in the location specified by output_folder before re-compiling.
* Add spline knots to the optimization result Adds spline knots to the optimization result so they can be easily obtained after optimization. This then allows easy further use, if needed. * Documentation update & fix notebook * Fix types * Daniel review change Co-authored-by: Daniel Weindl <dweindl@users.noreply.github.com> * add test spline knot saving * Update pypesto/visualize/spline_approximation.py Co-authored-by: Maren Philipps <55318391+m-philipps@users.noreply.github.com> --------- Co-authored-by: Daniel Weindl <dweindl@users.noreply.github.com> Co-authored-by: Maren Philipps <55318391+m-philipps@users.noreply.github.com>
* changed to new amici import format, removed some depracation warnings. Some are left, one needs to keep an eye out for probably. * corrected to amici.petab.conditions.fill_in_parameters * added minimal version to amici
* thermodynamic integration * compute_log_evidence * added constants in C.py * thermodynamic integration not supported for adaptive PT * adding logger for beta init
* added warnings to dynesty and support for posterior * adapted objective in test to likelihood * added string constants * added objective type to dynesty in notebook
Currently model simulation is run twice, even when no sensitivities are requested and the optimal inner parameters could be applied directly.
…#1328) Fixes an issue where ESSOptimizer would try to repeatedly perform a local search from the best known value, instead of trying other promising points, in case the initial local search yields a non-finite result.
`SacessOptimizer` is superior to `CESSOptimizer` and I wouldn't recommend using the latter. I think it's best to just remove it. No value, just extra maintenance effort. I don't think anybody was using it yet anyways, so this shouldn't actually be a breaking change for anybody.
…s` in favor of `DynestySampler.raw_results` (#1331)
* Fix return of inner parameters on objective call * hierarchical_decorator outside of history_decorator --------- Co-authored-by: Daniel Weindl <daniel.weindl@helmholtz-munich.de>
Codecov ReportAttention: Patch coverage is
❗ Your organization needs to install the Codecov GitHub app to enable full functionality. Additional details and impacted files@@ Coverage Diff @@
## main #1342 +/- ##
==========================================
- Coverage 84.54% 84.50% -0.04%
==========================================
Files 153 157 +4
Lines 12456 12917 +461
==========================================
+ Hits 10531 10916 +385
- Misses 1925 2001 +76 ☔ View full report in Codecov by Sentry. |
* First working example * temp fix * tmp_fix 2 * Added petab testsuite. Currently everything aside from Preequilibration and InitialAssignments overwrites works. * All except for one petab test working. I.e. functioning ss-computation * Sped up computation for MODE_FUN. Two tests only pass unreliably, one test expectedly fails. * Fixed error, where the conditions in edatas did not align with the parameter mappings * Removed pandas calculation to dataframe from calculator, made stand alone function. Expected speedup: ~5-10x * Added SolverOptions for roadrunner simulation and slightly more convenient way to interact with it. * Some cleanup * Added get_config. Skipping 0018 for now. * Support MODE_RES correctly * added petab testsuite to tox for roadrunner. * added roadrunner dependency * Case 0018 works now as well. Fixed a problem if not all observables were measured at all timepoints * Review suggestions and sanity check on filling parameters and edata * Removed "unscale_parameter" function. Considerably speed up model compilation. * removed timepoints from edatas. Added to documentation * Removed fill_replicates, as it is now possible with roadrunner. Also able to more conveniently setValues. Added newer version restriction to roadrunner. * Added roadrunner to tox.ini docs. * Added roadrunner to tox.ini base. * Changed to use API token for deployment * Added short docstring to each file * Added more descriptive docstring. * Reverted deploy changes, in favor of separate PR * Update pypesto/objective/roadrunner/roadrunner_calculator.py Co-authored-by: Daniel Weindl <dweindl@users.noreply.github.com> * Hopefully fixes doc build * Removed name for type for correct docstring. * Should fix docs now and also pandas warning * typo in tox * Integrated suggestions * Added roadrunner to doc installation --------- Co-authored-by: Daniel Weindl <dweindl@users.noreply.github.com>
Use 'spawn' by default, but allow passing other start methods via `SacessOptimizer(..., mp_start_method=...)`. Closes #1343.
Stop using deprecated type annotations and imports. Include ruff pyupgrade rules in pre-commit.
* Added deepcpy functionality and small test. * Better test and better deepcopy * Update pypesto/objective/roadrunner/road_runner.py Co-authored-by: Daniel Weindl <dweindl@users.noreply.github.com> * Removed deepcopy from base function as it should be similar to defaultbehaviour. * Also removed deepcopy functionality. * Update test/base/test_roadrunner.py Co-authored-by: Daniel Weindl <dweindl@users.noreply.github.com> --------- Co-authored-by: Daniel Weindl <dweindl@users.noreply.github.com>
* fix #1337 * fix pure callback calls * run test in 32bit and 64bit mode * add vmap compatibility, remove second order support * Update test_objective.py * Apply suggestions from code review Co-authored-by: Paul Jonas Jost <70631928+PaulJonasJost@users.noreply.github.com> * alter implementation, fix doc and extend tests * Update test_objective.py * expose inner x_names rather than allowing them to be set * Update base.py * Update test_objective.py --------- Co-authored-by: Paul Jonas Jost <70631928+PaulJonasJost@users.noreply.github.com> Co-authored-by: Daniel Weindl <daniel.weindl@helmholtz-munich.de>
* compute burn in for each chain * using all chains to compute evidence
* removed startpoint method from minimize as its depracated and per default in the problem. * Changed offset to one that should definitely catch the warning and test for warning. * When loading a problem, we expect a warning. * Trust-exact cannot handle bounds is an expected warning here. * If hess is not traced, no need to compute imo * Revert last changes, instead expect additional warning. * Better catching of warnings * Catch warnings? * Removed startpoint methods from functions, all were uniform. * added match as parameter to pytest.warns * Removed warning, is already tested elsewhere.
Merged
* Added a brief roadrunner example. to be extended. * Added roadrunner into tox.ini for notebook 1 * Filled in hyperrefs * Added missing hyperref --------- Co-authored-by: Doresic <85789271+Doresic@users.noreply.github.com>
* Add support for complex noise formulae * Removed comment * removed try...except in favor of if
PaulJonasJost
requested review from
stephanmg,
m-philipps,
dilpath,
dweindl,
FFroehlich,
Doresic and
vwiela
as code owners
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.
int(Ensembles: don't expectOptimizerResult.idto be convertible toint#1351)