Eigen vector Whitening

whitenning.py

import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from numpy import array
from numpy import mean
from numpy import cov
from numpy.linalg import eig, svd


path = 'drive/My Drive/FMNIST/'
x_train = pd.read_csv(path+'Train_Data.csv')
y_train = pd.read_csv(path+'Train_Labels.csv')

x_test = pd.read_csv(path+'Test_Data.csv')
y_test =  pd.read_csv(path+'Test_Labels.csv')

# merging train and test samples
x_train.columns = range(182)
x_test.columns = range(182)
X = pd.concat([x_train, x_test])


import matplotlib.pyplot as plt
import numpy as np
import pandas as pd
from numpy import array
from numpy import mean
from numpy import cov
from numpy.linalg import eig, svd

# X = np.array([[1,2],[2,4],[3,6]])
# X = pd.DataFrame(X)
X.plot.scatter(x=0,y=1)
plt.savefig("X.png")
print("================== Whitening using Eigen Values ===================")

M = mean(X.T, axis=1)
# center columns by subtracting column means
C = X - M
# calculate covariance matrix of centered matrix
X_cov = np.dot(C.T,C)/len(C)

# get the covariance matrix using numpy library
# X_cov = cov(X, rowvar=False, bias=True)

# eigenvalue decomposition of the covariance matrix
d, V = np.linalg.eigh(X_cov)

print("d=", d)
colinear = False
for v in d:  
  if v < 1E-5:
    colinear = True

if colinear:
  print("\nlinear correlatin was found\n\n")
else:
  print("\nNo linear correlatin was found\n\n")

D = np.diag(1. / np.sqrt(d+1E-18))

W = (D.dot(V.T))

X_white = (W.dot(C.T)).T

fig, ax = plt.subplots(figsize=(10,10))
plt.gca().set_aspect('equal', adjustable='box')
# X_white = pd.DataFrame(X_white)
# X_white.plot.scatter(x=0,y=1)

ax.scatter(X_white[:,0], X_white[:,1], c='g')
ax.set_title("Whitened")
fig.savefig('w.png')