Releases: acerbilab/pybads
Fix depreacted Nans
What's changed
- Removed deprecated usage of np.NaN and np.Inf
- Removed unused package cmae
Full change log: v1.0.4...v1.0.5
PyBADS JOSS release
What's changed
- Official JOSS version of PyBADS
- Add checks on input bounds and change message errors
- New descriptions of the documentation
Full change log: v1.0.3...v1.0.4
PyBADS v1.0.3
What's changed
- Add checks and messages logs for the transformed gridized initial point and non-box constraints violations
Full change log: v1.0.2...v1.0.3
PyBADS v1.0.2
What's changed
- Minor fixes for pytest warnings and fixed minor bug for 1D optimization
- Moved optional dependencies to dev env
Full change log: v1.0.1...v1.0.2
PyBADS v1.0.1
- Fix bug wrong reshape when broadcasting a matrix operation for high-dimension problems (
$D > 32$ ) during the evolutionary search.
PyBADS v1.0.0
First full-version release of PyBADS, a Python package for fast and robust black-box optimization. Full documentation is available at https://acerbilab.github.io/pybads/. Feedback is welcome, see troubleshooting and contact. The same packaged version is also available at https://pypi.org/project/PyBADS/#history.
Additional detail of the algorithm can be found the paper Practical Bayesian Optimization for Model Fitting with Bayesian Adaptive Direct Search paper published in NeurIPS in 2017.
A MATLAB implementation is also available at the acerbilab/BADS repository.
PyBADS public beta v0.8.2 - conda release
Initial public beta version of PyBADS, a Python package for solving difficult and moderately expensive optimization problems. Full documentation is available at acerbilab.github.io/pybads/. Feedback is welcome, see troubleshooting and contact. The same packaged version is also available at pypi.
Additional details of the algorithm can be found in the original Bayesian Adaptive Direct Search paper presented at NeurIPS in 2017.
A MATLAB implementation is also available at the acerbilab/BADS repository.
PyBADS public beta v0.8.1
Initial public beta version of PyBADS, a Python package for solving difficult and moderately expensive optimization problems. Full documentation is available at acerbilab.github.io/pybads/. Feedback is welcome, see troubleshooting and contact. The same packaged version is also available at pypi.
Additional details of the algorithm can be found in the original Bayesian Adaptive Direct Search paper presented at NeurIPS in 2017.
A MATLAB implementation is also available at the acerbilab/BADS repository.
PyBADS public beta
Initial public beta version of PyBADS, a Python package for solving difficult and moderately expensive optimization problems. Full documentation is available at acerbilab.github.io/pybads/. Feedback is welcome, see troubleshooting and contact. The same packaged version is also available at pypi.
Additional details of the algorithm can be found in the original Bayesian Adaptive Direct Search paper presented at NeurIPS in 2017.
A MATLAB implementation is also available at the acerbilab/BADS repository.