Releases: relf/egobox
0.29.1
0.29.0
What's Changed
- Add experimental Global Bounded Nelder-Mead infill criterion optimizer by @relf in #264
- Add Rust simple examples by @relf in #266
- Adjust bounds for GP/SGP theta hyperparameters optimization by @relf in #258
- Make GP max nb of likelihood evaluations in hyperparameter optimization configurable by @relf in #262
- Make CoEGO activity matrix more balanced by @relf in #265
- Fix MOE smooth derivatives test which fails from time to time by @relf in #260
- Refactor infill criterion optimization by @relf in #259
- Remove last resort LHS optimization by @relf in #261
Full Changelog: 0.28.1...0.29.0
0.28.1
0.28.0
What's Changed
This release contains an implementation of the CoEGO [1] algorithm in a CCBO [2] setting where expensive objective evaluations are done only when all partial optimization problems are solved.
- Zhan, Dawei, et al., A cooperative approach to efficient global optimization. Journal of Global Optimization 88.2 (2024): 327-357
- Lisa Pretsch et al., Bayesian optimization of cooperative components for multi-stage aero-structural compressor blade design. 2024
See also some breaking changes description in optimizer options in PR #253
Full Changelog: 0.27.1...0.28.0
0.27.1
What's Changed
- Fix theta optim decision using q_points iter nb by @relf in #248
- Fix probability of feasibility product gradient by @relf in #249
- Fix constraint handling and prepare CoEGO implementation by @relf in #251
- Fix CI release actions by @relf in #247
- Fix links to crates by @jonaspleyer in #250
New Contributors
- @jonaspleyer made their first contribution in #250
Full Changelog: 0.27.0...0.27.1
0.27.0
0.26.0
What's Changed
See this example of an optimization with constraint functions compared to the same example where the constraints are provided with the objective (the legacy way) and are approximated with surrogates behind the scene. Constraint functions definition is suited when constraint computation is cheap to avoid surrogate approximation.
To summarize, Egor
handles three types of constraint:
- the ones defined by the bounds of the input domain (hypercube),
- the ones defined with the objective function which are approximated with surrogates
- the ones defined as separate cheap-to-compute functions
Notes:
- Gradient definition of a constraint function is required when using SLSQP as infill optimizer
- Constraint tolerance is defined as a float list of length (nb of surrogated constraints p (cstr) + nb of constraints function q (fcstr)) following the order of definition: [cstr1, ... cstr_p, fcstr_1, ..., fcstr_q]
Full Changelog: 0.25.2...0.26.0
0.25.2
What's Changed
- Fix python action by @relf in #229
- Fix working directory in python release action by @relf in #230
- Add readme specific to python egobox binding by @relf in #231
- Re-enable BLAS backend tests by @relf in #234
- Add stub file (type annotations) for Python APIs by @unkcpz in #235
- Fix Python example optim g24 by @relf in #238
New Contributors
Full Changelog: 0.25.1...0.25.2
0.25.1.post1
What's Changed
Full Changelog: 0.25.1...0.25.1.post1