New Toolbox of Broad Learning System, with sklearn liked APIs. Created at 2023.08
-
《A survey of current Broad learning Models》- (English Version, South China University of Technology) : https://ieeexplore.ieee.org/abstract/document/9380770/
-
《宽度学习研究进展》-(中文版,华南理工大学) : http://www.cnki.com.cn/Article/CJFDTotal-JSYJ202108003.htm
-
BLS(South China University of Technology, 2018) : Based Broad Learning System, including BLSRegressor and BLSClassifier. API
fit
andpredict
is available now.GridSearch
will online soon. Paper Link: https://ieeexplore.ieee.org/abstract/document/7987745/ -
Stacked-BLS(South China University of Technology, 2021) : A Stacked Model of Broad Learning System, with several residual learning block, including StackedBLSClassifier. API
fit
andpredict
is available now. Paper Link: https://ieeexplore.ieee.org/abstract/document/9308927/ -
BLS-AutoEncoder(South China University of Technology, 2023) : A novel Auto-Encode structure based on BLS, which is used to extract the feacture of original data. These APIs contains BLSAEExtractor and StackedBLSAEExtractor. API
fit
is available now, whose output is feature vectors, not prediction. Please check the usage of APIs intest_demo.py
Paper Link: https://ieeexplore.ieee.org/abstract/document/9661311 -
BLS-LRF : An ensemble model based on a simple convoloution module and a Broad Learning System module, which enhance the ability of feature extraction for original BLS. API of this model would be upload sonn.
-
Deep-Broad Ensemble Model(JiNan University, 2023) : A novel model based on 3D-Residual Conv Module and Broad Learning System. Upload soon. Paper Link: https://www.frontiersin.org/articles/10.3389/fnins.2023.1137557/full
-
Tree-BLS(Beijing University of Technology, 2022) : A novel BLS model for enhancing the efficiency of small data modeling with various dimensions, whose original mapping neurons is replaced by tree modules. Paper Link: https://ieeexplore.ieee.org/abstract/document/9938406