tmorshed / padasip Goto Github PK

This library is designed to simplify adaptive signal processing tasks within python (filtering, prediction, reconstruction, classification). For code optimisation, this library uses numpy for array operations.

Also in this library is presented some new methods for adaptive signal processing. The library is designed to be used with datasets and also with real-time measuring (sample-after-sample feeding).

Everything is on github:

http://matousc89.github.io/padasip/

• Principal Component Analysis (PCA)
• Linear Discriminant Analysis (LDA)

The library features multiple adaptive filters. Input vectors for filters can be constructed manually or with the assistance of included functions. So far it is possible to use following filters:

• LMS (least-mean-squares) adaptive filter
• NLMS (normalized least-mean-squares) adaptive filter
• LMF (least-mean-fourth) adaptive filter
• NLMF (normalized least-mean-fourth) adaptive filter
• SSLMS (sign-sign least-mean-squares) adaptive filter
• NSSLMS (normalized sign-sign least-mean-squares) adaptive filter
• RLS (recursive-least-squares) adaptive filter
• GNGD (generalized normalized gradient descent) adaptive filter
• AP (affine projection) adaptive filter

So far it is implemented only MLP neural network.

The library features two novelty/outlier detection tools

• Error and Learning Based Novelty Detection (ELBND)
• Learning Entropy (LE)