Spectral analysis
In the previous section, we charted the amplitude spectrum of the dataset. The power spectrum of a physical signal visualizes the energy distribution of the signal. We can modify the code easily to plot the power spectrum, just by squaring the values as follows:
plt.plot(transformed ** 2, label="Power Spectrum")
The phase spectrum visualizes the phase (the initial angle of a sine function) and can be plotted as follows:
plt.plot(np.angle(transformed), label="Phase Spectrum")
Refer to the following graph for the end result:

Please refer to the ch-07.ipynb
file in this book's code bundle for the complete code.