Periodogram.py

# -*- coding: utf-8 -*-

# -*- coding: utf-8 -*-

import numpy as np
import matplotlib.pyplot as plt
from scipy.fftpack import fft
from IPython import get_ipython
import seaborn as sns
import glob
import pandas as pd
from scipy import signal
#from sklearn.preprocessing import MinMaxScaler
from sklearn.preprocessing import StandardScaler
from sklearn.preprocessing import Normalizer


# extracted from : https://stackoverflow.com/questions/39032325/python-high-pass-filter
def butter_highpass(cutoff, fs, order=5):
    nyq = 0.5 * fs
    normal_cutoff = cutoff / nyq
    b, a = signal.butter(order, normal_cutoff, btype='high', analog=False)
    return b, a

def butter_highpass_filter(data, cutoff, fs, order=5):
    b, a = butter_highpass(cutoff, fs, order=order)
    y = signal.filtfilt(b, a, data)
    return y


def psd(node):
    y = df[node]
    y = butter_highpass_filter(y,5,132,5)   
    f, psda= signal.periodogram(y,256)
    return f,psda



file_ = ('./Data/ssvep/dinuka_8' + ".csv")

df = pd.read_csv(file_,index_col=None, header=0,skiprows=range(1, 35))

rows,clmns = df.shape 

df2=df
df3=df[1:5]
df = df.loc[128 :5*128]

# specifying the O2 node for the value

f1,p1 = psd('O1 Value')
f2,p2 = psd('O2 Value')
#f3,p3,i3 = fourier('P7 Value')
#f4,p4,i4 = fourier('P8 Value')
#f5,p5,i5 = fourier('F3 Value')


plt.figure(1)
plt.subplot(211)
plt.plot(f2 , p2,label="O2")
plt.plot(f1 , p1,label="O1")
#plt.plot(f3[i3] , p3[i3],label="P7")
#plt.plot(f4[i4] , p4[i4],label="P8")
#plt.plot(f5[i5] , p5[i5],label="F3")

plt.legend(bbox_to_anchor=(0., 1.02, 1., .102), loc=3,
           ncol=2, mode="expand", borderaxespad=0.)


plt.show()