Nature-inspired algorithms for hyper-parameter tuning of Scikit-Learn models.
MIT License
Bot releases are visible (Hide)
Published by github-actions[bot] about 1 year ago
Published by github-actions[bot] almost 2 years ago
Support for Python 3.11.
Full Changelog: https://github.com/timzatko/Sklearn-Nature-Inspired-Algorithms/compare/v0.10.0...v0.11.0
Published by github-actions[bot] almost 2 years ago
Update niapy to version v2.0.4.
Full Changelog: https://github.com/timzatko/Sklearn-Nature-Inspired-Algorithms/compare/v0.9.0...v0.10.0
Published by github-actions[bot] over 2 years ago
Update niapy
to version v2.0.2
.
Published by github-actions[bot] over 2 years ago
Fix cross-validation could not be set via object (using GroupKFold
for example).
Published by github-actions[bot] over 2 years ago
Add support for python 3.10.x and drop support for 3.7.x.
Published by github-actions[bot] almost 3 years ago
Published by github-actions[bot] almost 3 years ago
Update niapy
package version to v2.0.0.
Published by github-actions[bot] about 3 years ago
Update niapy
package version to 2.0.0rc18
.
Published by github-actions[bot] about 3 years ago
Update niapy
package to v2.0.0rc17
(renamed from NiaPy
).
Published by github-actions[bot] over 3 years ago
Fix scikit-learn deprecation warning.
Published by github-actions[bot] almost 4 years ago
Fixed minimum versions of dependencies.
Published by github-actions[bot] almost 4 years ago
Updated dependencies.
From:
toml = "^0.9"
scikit-learn = "^0.22.2"
numpy = "^1.18.4"
matplotlib = "^3.2.1"
seaborn = "^0.10.1"
pandas = "^1.0.3"
To:
toml = "^0.10"
scikit-learn = "^0.23"
numpy = "^1.19"
matplotlib = "^3.3"
seaborn = "^0.11"
pandas = "^1.1"
Published by github-actions[bot] about 4 years ago
Updated documentation.
Published by github-actions[bot] over 4 years ago
Better logging.
Published by github-actions[bot] over 4 years ago
Added documentation via http://readthedocs.io/.
Published by github-actions[bot] over 4 years ago
Updated README with more examples.
Published by github-actions[bot] over 4 years ago
random_state
option NatureInspiredSearchCV
Published by github-actions[bot] over 4 years ago
The nature-inspired search can be run multiple times, this will provide more stable results. Helper visualization functions were updated to show multiple runs/select run to show.
Published by github-actions[bot] over 4 years ago
NatureInspiredSearchCV
, the new defaults are:
algorithm
= hba
population_size
= 50
max_n_gen
= 100
max_stagnating_gen
= 20