A hyperparameter optimization toolbox for convenient and fast prototyping
A hyperparameter optimization and meta-learning toolbox for convenient and fast prototyping of machine-learning models.
Hyperactive is primarly a hyperparameter optimization toolkit, that aims to simplify the model-selection and -tuning process. You can use any machine- or deep-learning package and it is not necessary to learn new syntax. Hyperactive offers high versatility in model optimization because of two characteristics:
- You can define any kind of model in the objective function. It just has to return a score/metric that gets maximized.
- The search space accepts not just int, float or str as data types but even functions, classes or any python objects.
For more information, visualization and details about the API check out the
- Thoroughly tested code base
- Compatible with any python machine-learning framework
- Utilize state of the art optimization techniques like:
- Simulated annealing
- Evolution strategy
- Bayesian optimization
- High performance: Optimizer time is neglectable for most models
- Choose from a variety of different optimization extensions to improve the optimization
|Optimization Techniques||Tested and Supported Packages||Optimization Extensions|
|Local Search: Random Methods: Markov Chain Monte Carlo: Population Methods: Sequential Methods:||Machine Learning: Deep Learning: Distribution:||
The most recent version of Hyperactive is available on PyPi:
pip install hyperactive
- [x] Change API
- [x] Ray integration
- [x] Save memory of evaluations for later runs (long term memory)
- [x] Warm start sequence based optimizers with long term memory
- [x] Gaussian process regressors from various packages (gpy, sklearn, GPflow, ...) via wrapper
- [x] Add basic dataset meta-features to long term memory
- [x] Add helper-functions for memory
- [x] connect two different model/dataset hashes
- [x] split two different model/dataset hashes
- [x] delete memory of model/dataset
- [x] return best known model for dataset
- [x] return search space for best model
- [x] return best parameter for best model
- [ ] Tree-structured Parzen Estimator
- [ ] Multiple experimental SBOM
- [ ] Meta-Optimization of all optimizers
- [ ] Spiral optimization
- [ ] Downhill-Simplex-Method
- [ ] Differential evolution
- [ ] Helper-functions for early stopping
- [ ] Helper-functions for dataset approximation
The following algorithms are of my own design and, to my knowledge, do not yet exist in the technical literature. If any of these algorithms already exist I would like you to share it with me in an issue.
A combination between simulated annealing and random search.
Inspired by hyperband optimization.
Release history Release notifications
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
|Filename, size||File type||Python version||Upload date||Hashes|
|Filename, size hyperactive-2.2.2-py3-none-any.whl (47.6 kB)||File type Wheel||Python version py3||Upload date||Hashes View hashes|
Hashes for hyperactive-2.2.2-py3-none-any.whl