import numpy
from spectrum import speriodogram, data_cosine
from pylab import figure, semilogy, figure ,imshow
# create N data sets and make the frequency dependent on the time
N = 100
m = numpy.concatenate([data_cosine(N=1024, A=0.1, sampling=1024, freq=x) 
    for x in range(1, N)]);
m.resize(N, 1024)
res = speriodogram(m)
figure(1)
semilogy(res)
figure(2)
imshow(res.transpose(), aspect='auto')