Releases: thieu1995/MetaPerceptron
Releases · thieu1995/MetaPerceptron
v2.2.0
v2.1.0
- Fix bug name of optimizer in 
MhaMlpmodels - Fix bug encoding labels data in 
Dataclass - Add GPU support for Gradient Descent-trained MLP models
 - Rename 
optim_parastooptim_paramsin all models. - Fix bug in 
validatormodule - Fix bug missing loss_train in Gradient Descent-trained MLP models
 - Update examples, tests, documents, citations, and readme.
 
v2.0.0
- Re-structure and re-implement the MLP, Metaheuristic-trained MLP, and Gradient Descent-trained MLP models.
 - Remove dependency on the Skorch library.
 - Add the 
MhaMlpTunerclass, which tunes hyperparameters for Metaheuristic-based MLP models. - Add the 
MhaMlpComparatorclass, which compares multiple Metaheuristic-based MLP models. - Update examples, tests, and documentation.
 - Update requirements, citations, and README.
 
v1.1.0
v1.0.1
- Add infors (CODE_OF_CONDUCT.md, MANIFEST.in, LICENSE, requirements.txt)
 - Add helpers modules (act_util, metric_util, scaler_util, validator, preprocessor)
 - Add MlpRegressor and MlpClassifier classes (Based on Pytorch and Skorch)
 - Add publish workflow
 - Add examples and tests folders