Discrete Wavelet Transform (DWT).
Initialize the Dwt class.
Input
- specdiff - [string] spectral difference method (‘rpv’, ‘arpv’, ‘crpv’)
Return ABSOLUTE feature weights.
Input
- x - [2D numpy array float] (sample x feature) training data
- y - [1D numpy array integer] (two classes, 1 and -1) classes
Output
- fw - [1D numpy array float] feature weights
[Subramani06] | P Subramani, R Sahu and S Verma. Feature selection using Haar wavelet power spectrum. In BMC Bioinformatics 2006, 7:432. |