Skip to content

Latest commit

 

History

History
40 lines (25 loc) · 865 Bytes

README.md

File metadata and controls

40 lines (25 loc) · 865 Bytes

RVFLN

A Python implementation of Random Vector Functional Link Networks and Broad Learning Systems using the Sklearn API

Installation

pip install rvfln

Usage

Since the API is based on that of sklearn the usage is very similar. The package contains:

  • An rvfln module containing a regressor and a classifier estimator
  • A bls module containing a regressor and a classifier estimator

Example:


from rvfln.bls import BLSClassifier

clf = BLSClassifier(
    n_z = 10,
    n_z_features = 400,
    n_h = 2000,
    alpha = 1
).fit(x_train, y_train)

print(clf.score(x_test, y_test))

YOH-HAN PAO, STEPHEN M. PHILLIPS & DEJAN J. SOBAJIC (1992) Neural-net computing and the intelligent control of systems, International Journal of Control, 56:2, 263-289, DOI: 10.1080/00207179208934315