Simon Blanke

Results 55 issues of Simon Blanke

We have a problem: https://github.com/SimonBlanke/Hyperactive/actions/runs/15483542700/job/43593595699#step:5:53 Error: from sklearn.utils.deprecation import _deprecate_Xt_in_inverse_transform ImportError: cannot import name '_deprecate_Xt_in_inverse_transform' from 'sklearn.utils.deprecation' This is what sklearn says about this: https://github.com/scikit-learn/scikit-learn/issues/29366 I had this error before:...

bug

Hyperband is a hyperparameter optimization algorithm that improves efficiency by dynamically allocating computational resources. It is based on the Successive Halving Algorithm (SHA) but introduces an adaptive mechanism to balance...

enhancement

To separate the discussion about the optimization-strategy api for v5 started in https://github.com/SimonBlanke/Hyperactive/pull/101. In the future it makes sense to rename this feature to something like "CompositeOptimizer" or "OptimizerPipeline". In...

This would add the harmonic search algorithm to the collection of optimizers. I was not successful in accessing the original paper: "Geem, Z. W., Kim, J. H., & Loganathan, G....

enhancement

Implement the COBYLA algorithm, a gradient-free method for constrained nonlinear optimization. It models both the objective and constraint functions with linear approximations and uses a trust-region approach. Suitable for problems...

enhancement

Implement the UOBYQA algorithm, which constructs a quadratic approximation of the objective function using interpolation at a set of points. It is a derivative-free method for unconstrained optimization. For information...

enhancement

Implement NEWUOA, an improved derivative-free optimization method for unconstrained problems. It builds quadratic models of the objective function using a well-chosen set of interpolation points. For information about the mathematical...

enhancement

Implement the BOBYQA algorithm, a derivative-free optimization method for problems with bound constraints. It extends NEWUOA by accommodating box constraints on the variables, while still employing a quadratic approximation strategy...

enhancement

This aims to add a replacement parameter to the random search optimization. From what I can tell, this would completely change the way random search works. So the random search...

enhancement