def make_feature_vector(readings): # A function we apply to each group of power spectra
'''
Create 100, log10-spaced bins for each power spectrum.
For more on how this particular implementation works, see:
http://coolworld.me/pre-processing-EEG-consumer-devices/
'''
return brainlib.avgPowerSpectrum(brainlib.binnedPowerSpectra(spectra(readings), 100), np.log10)
def make_feature_vector (readings): # A function we apply to each group of power spectra
'''
Create 100, log10-spaced bins for each power spectrum.
For more on how this particular implementation works, see:
http://coolworld.me/pre-processing-EEG-consumer-devices/
'''
return brainlib.avgPowerSpectrum(
brainlib.binnedPowerSpectra(spectra(readings), 100)
, np.log10)
def bin_power_spectrum(power_spectrum):
'''Create 100, log10-spaced bins for each power spectrum. See the work of Merrill et al.:
Merrill, N., Maillart, T., Johnson, B., & Chuang, J. "Improving Physiological Signal
Classification Using Logarithmic Quantization and a Progressive Calibration Technique."
Physiological Computing Systems 2015. Angers, France.'''
return(brainlib.binnedPowerSpectra(power_spectrum, 100))