def psd(datapdf): d = data.read(result["path"]) nfft = 256 * 2 data2anz = result["attenddata"] pow = zeros([(nfft / 2) + 1, size(data2anz, 1)]) comp = zeros([(nfft / 2) + 1, size(data2anz, 1)], complex_) # fftreal = (zeros([(nfft/2)+1, size(attenddata,1)])) for eachch in range(0, size(data2anz, 1)): p, f, i = spectral.psd( data2anz[:, eachch], NFFT=nfft, Fs=1 / d.hdr.header_data.sample_period ) # , noverlap=200)#, detrend=matplotlib.pylab.detrend_mean) # p,f,i = spectral.psd(data[:,eachch], NFFT=nfft, Fs=1/ d.hdr.header_data.sample_period ) pow[:, eachch] = p comp[:, eachch] = i # fftreal[:,eachch] = r logpow = 10 * log10(pow) freq = f
def psd(datapdf): d = data.read(datapdf) nfft=256*2 ch = channel.index(datapdf, 'meg') d.getdata(0, d.pnts_in_file, ch.channelindexhdr) data2anz = d.data_block #data2anz = result['attenddata'] pow = zeros([(nfft/2)+1, size(data2anz,1)]) comp = zeros([(nfft/2)+1, size(data2anz,1)], complex_) #fftreal = (zeros([(nfft/2)+1, size(attenddata,1)])) for eachch in range(0, size(data2anz,1)): p,f,i = spectral.psd(data2anz[:,eachch], NFFT=nfft, Fs=1/ d.hdr.header_data.sample_period)#, noverlap=200)#, detrend=matplotlib.pylab.detrend_mean) #p,f,i = spectral.psd(data[:,eachch], NFFT=nfft, Fs=1/ d.hdr.header_data.sample_period ) pow[:,eachch] = p comp[:,eachch] = i #fftreal[:,eachch] = r logpow = 10*log10(pow) freq=f return logpow, freq