Abstract Differential Evolution (DE) Sampler This implementation introduces a novel Differential Evolution (DE) sampler, tailored to optimize both numerical and categorical hyperparameters effectively. The DE sampler integrates a hybrid approach:
Differential Evolution for Numerical Parameters: Exploiting DE’s strengths, the sampler efficiently explores numerical parameter spaces through mutation, crossover, and selection mechanisms. Random Sampling for Categorical Parameters: For categorical variables, the sampler employs random sampling, ensuring comprehensive coverage of discrete spaces. The sampler also supports dynamic search spaces, enabling seamless adaptation to varying parameter dimensions during optimization.
Class or Function Names DEHBSampler DEHBPruner Installation There is no additional installation required for this sampler and pruner, but if you want to run the example.py script, you need to install the following packages:
$ pip install sklearn Example sampler = DEHBSampler() pruner = DEHBPruner(min_resource=1, max_resource=n_train_iter, reduction_factor=3) study = optuna.create_study(sampler=sampler, pruner=pruner) See example.py for a full example. The following figures are obtained from the analysis of the optimization. Others References Awad, N.