<summary> perform data downsampling, it just aggregates data without applying lowpass filter </summary> <param name="data"></param> <param name="perio

perform data downsampling, it doesnt apply lowpass filter for you, it just aggregates several data points :param data: initial data :

<summary> perform direct fft </summary> <param name="data">data for fft</param> <param name="start_pos">start pos</param> <param name="end_pos">end po

perform direct fft :param data: data for fft, len of data must be even :type data: NDArray[Shape["*"], Float64] :param window

Apply high pass filter to provided data.

<summary> perform highpass filter, unlike other bindings instead in-place calculation it returns new array </summary> <param name="data"></param> <par

apply high pass filter to provided data :param data: data to filter, filter works in-place :type data: NDArray[Shape["*"], Float64]

<summary> Calculate ICA </summary> <param name="data"></param> <param name="num_components"></param> <returns></returns>

perform ICA :param data: 2d array for calculation :type data: NDArray[Shape["*, *"], Float64] :param num_components: number o

<summary> perform inverse fft </summary> <param name="data">data from perform_fft</param> <returns>restored data</returns>

perform inverse fft :param data: data from fft :type data: NDArray[Shape["*"], Complex128] :return: restored data :rt

<summary> perform inverse wavelet transorm </summary> <param name="wavelet_data">tuple returned by perform_wavelet_transform</param> <param name="orig

perform wavelet transform :param wavelet_output: tuple of wavelet_coeffs and array with lengths :type wavelet_coeffs: tuple :

<summary> perform lowpass filter, unlike other bindings instead in-place calculation it returns new array </summary> <param name="data"></param> <para

apply low pass filter to provided data :param data: data to filter, filter works in-place :type data: NDArray[Shape["*"], Float64]

<summary> perform moving average or moving median filter, unlike other bindings instead in-place calculation it returns new array </summary> <param na

smooth data using moving average or median :param data: data to smooth, it works in-place :type data: NDArray[Shape["*"], Float64]

<summary> perform wavelet based denoising </summary> <param name="data">data for denoising</param> <param name="wavelet">use WaveletTypes enum</param>

perform wavelet denoising :param data: data to denoise :type data: NDArray[Shape["*"], Float64] :param wavelet: use WaveletTy

<summary> perform wavelet transform </summary> <param name="data">data for wavelet transform</param> <param name="wavelet">use WaveletTypes enum</para

perform wavelet transform :param data: initial data :type data: NDArray[Shape["*"], Float64] :param wavelet: use WaveletTypes

Instead of allocating a special handle, the pipe_*_new() functions just return the original pipe, cast into a user-friendly form. This saves needless

<summary> Get score of classifier </summary>

Calculate metric from data.

<summary> Prepare classifier </summary>